Over the last few months, I’ve used ChatGPT’s deep research feature, and while I don’t agree with those who claim it’s as good at research or writing as most graduate students, it's certainly a useful tool for me. I believe many educators would find it helpful when looking for research on various educational strategies or for the latest findings in their field of study.

Like many, I’ve increasingly begun to use AI as an alternative to a traditional search engine. Deep Research works as a beefed-up version of that. It can be quicker and more efficient than searching a site such as Google Scholar, and can help you find different results than you might find searching more traditional academic archives.

Here’s a closer look at how OpenAI’s Deep Research works, and how it performed on various academic prompts with which I tested it.

How ChatGPT’s Deep Research Works

Deep research is designed to work independently by finding, analyzing, and summarizing hundreds of online searches. The tool is powered by OpenAI’s o3 model and uses reasoning—meaning it can break down problems into smaller components, make predictions, infer information, and apply logic.

Those who use the free version of ChatGPT, get five free deep research prompts per month. Plus, Team, Enterprise, and Edu users get 25 per month, while Pro subscribers get 250 prompts.

To access Deep Research, you need to click on the tools icon at the bottom of the chatbot used to chat with ChatGPT. Then select “run deep research.”

After you put in a prompt, the tool will ask several clarifying questions about the type of research you want. Once you’ve answered these, it will take a few minutes to generate a response.

Here are some of the prompts I used and what I thought about the results Deep Research generated.

Flipped Learning Prompt

The Prompt: Can you write a summary of the research on flipped learning? Pay particular attention to randomized control trials looking at the educational strategy in college and k-12 settings.

The Result: ChatGPT quickly responded with a detailed summary of the research into flipped learning with effects divided between college and K-12 research. The summary concluded the intervention had a moderate but fairly consistent effect observed in most studies. ChatGPT provided links to pertinent research with which I was already familiar, but which had taken me quite a while to track down. Additionally, ChatGPT provided several interesting studies I hadn’t come across.

Impressively, the overview also described some of the limitations around flipped learning, addressing how it can increase the workload outside of class and how teachers who attempt to implement it can find significant pushback from students. ChatGPT also cited a 2019 paper describing a randomized control trial looking at flipped learning conducted at West Point that found short-term gains in Math and no effect in Economics. The study also suggested that the intervention could widen the achievement gap: as positive effects were driven by white male, and higher-achieving students.

The study is an outlier, so it's hard to say what its significance is, but it was interesting to see and an important part of the flipped learning conversation.

Learning Styles Prompt

The Prompt: Is there any evidence that learning styles are real that you can find in academic journals?

The Result: ChatGPT did a nice job pointing out the problems with this approach to education, writing: “Major meta-analyses and reviews – especially in the last 15 years – largely refute the learning styles hypothesis, or at best find only weak effects. The current consensus in the academic community is that educators should be skeptical of the learning styles approach.”

The overview it generated also included many studies that supported this conclusion. Because of the way my prompt was phrased, I think it encouraged ChatGPT to look for evidence in support of this theory. It found some older studies that suggested a link, but was quick to point out the problems with this earlier research.

I was impressed with ChatGPT’s response. I’ve written about the persistent myth of learning styles before, and worried ChatGPT’s research mode would be confused by all the online chatter from educators in support of this debunked theory. It wasn’t.

Writing and Cognition Prompt

The Prompt: Can you research the potential positive link between writing and thinking/cognition?

The Result: This is a topic I’m curious about as a writer and writing professor, but I’ve only come across a few studies and haven’t yet done a deep research dive myself. ChatGPT may have saved me the effort.

The response here provided a wide-ranging and well-cited overview of different ways in which writing can help our thinking processes, from increasing memory to reducing stress and improving critical thinking and metacognition.

One example of the type of research included in the summary is a fascinating study that found college freshmen who wrote about the stress of starting college saw larger working memory improvements than their classmates who did not have this writing assignment.

Conclusion

Deep Research is one of the more helpful AI tools I’ve come across for academics, and I believe it can be helpful to educators in various fields and levels. By essentially generating high-quality Wikipedia articles on demand, it’s a great search tool that can help make research more efficient. I’ve found it helpful in the examples above and in other uses.

However, much like Wikipedia itself, it’s a good place to start your research and a bad place to end it. Deep Research can help you get a fast overview of a topic and suggest research to review, but actually assessing that research and deciding what implications it has is still best performed by a human.

• DeepSeek: Everything Educators Need to Know About The New AI Model
• AI In Education: 4 Ways To Avoid Falling Behind The Curve

While some schools have started to introduce AI literacy requirements, it is not something that is widespread. Moreover, students (and teachers) need to understand not only what AI is and how it functions, but also how it impacts many facets of our society, and its potential for harm.

In this article, we share free lessons and resources that you can incorporate into lessons for students about using AI.

AI Overview

For an overall understanding of AI, Code.org has a wonderful set of resources, including lessons that are aligned with computer science standards as well as concise-but-clear videos that break down related topics. The instructional videos are under the “How AI Works” program and explore topics on machine learning, training and data bias, neutral networks, computer vision, chatbots, and large language models. Learning materials are connected to the videos that can be used in lessons.

1. AI Responsible and Ethical Use

While Code.org briefly highlights biases that are inherent in AI, it is important to intentionally expand on AI responsible and ethical use from a tech justice lens. The Kapor Center offers a free Responsible AI and Tech Justice Guide for K-12 education. The guide is organized into six core components with the aim of building students’ capacity to explore the impacts and implications of AI technologies on society, be aware of personal usage of AI, and examine the AI creation ecosystem, among other goals. The comprehensive guide is packed with lessons, case studies, articles, and other resources.

2. Al Environmental Impacts

AI has major impacts on the environment, agriculture, water consumption, energy, and more, so there are developed lessons that you can use to teach students about these concerns. You can also build a focus on sustainability and taking personal actions to mitigate some of AI’s carbon footprint. Partially funded by the National Science Foundation (NSF), this Environmental Impact of AI Lesson, which is under a Creative Common license, can be used. It features a vocabulary review, an interactive activity, and a planning guide as well as a section on teacher modifications that include teacher-made resources aligned with the lesson.

Similarly, the Share My Lesson community created by the American Federation of Teachers (AFT) offers “AI’s Environmental Impact: The Growing Strain on Power Grids and Local Communities.” This resource includes video, defined key terms, discussion questions, extension content, and suggested deliverables.

3. AI Misinformation

The News Literacy Project offers Checkology, a free virtual platform with a curriculum to develop students’ ability to confirm when information is from credible sources, understand what a credible source is and where to locate one, and to use critical thinking to assess the credibility of content. The Checkology platform has a host of resources and interactive activities for students to engage in while learning about misinformation. It also has an evaluation center for teachers to track student progress and a help center for support. The lessons within the platform are aligned with myriad standards, including ISTE Student Standards, Next Generation Science Standards, Common Core ELA Standards, American Association of School Librarians Standards Framework for Learners, and the College, Career, and Civic Life (C3) Framework for Social Studies State Standards.

4. AI Deepfakes

AI for Education provides downloadable resources that include a classroom guide and discussion questions to use as you teach students about misinformation and AI deepfakes. The materials will help students understand what deepfakes are, the dangers of deepfakes, such as cyberbullying and misinformation, and techniques that are used to create deepfakes. It also offers real examples of deepfakes that feature popular artists and books, such as Katy Perry and Harry Potter.

Which Face is Real? also offers a game that you could play with students after they learn about deepfakes. In the game, two images appear (one AI-generated and one real image). The students then need to apply the skills they have learned on misinformation and deepfakes, and choose the one that is real. Which Face is Real? also has a Learn tab with information that can be shared with students as well.

We hope some of these resources are helpful as you prepare your students to understand and responsibly use AI. If you feel like you would like training yourself first, check out these 10 Free Certifications and Trainings for Educators.

• Developing AI Pedagogical Practices
• AI Lesson Plan Tips
• Explaining AI in the Classroom to Families

The current version of Kira offers AI tools that are built to support teachers and students across educational scenarios and subjects. The company says this gives teachers more time to focus on students and classroom impact.

Built by a team of educators, engineers, designers, and operators, Kira is a multi-faceted system that offers a finished and complete AI solution for teachers' needs, and is free for educators and students.

The company claims, "Kira is the first AI platform that delivers personalized education at scale."

This guide lays out all you need to know about Kira in your class.

What is Kira Learning?

Kira Learning is an AI platform that works to make the job of teachers more simple while also supporting students in their learning.

This is actually described by its creators as an AI toolkit consisting of more than 20 different AI tools, so it can cover a wide variety of teacher and student needs.

This can be used to prepare, with lesson plans and rubrics; to assess with quiz design and differentiated instructions; and to grade with feedback generation, progress tracking, and more.

A course library for teaching computer science is also available -- listed below -- since this does still offer the original Kira content, only now with a way more AI-focused approach.

How does Kira Learning work?

Kira Learning offers a rich course catalog that you can get into for computer science teaching. But now there is all that AI power as well, it's possible to use it in a way that is focused on more general teaching.

Sign up with a Google or school account initially and as a teacher you can begin using the services. You build activities using AI, create a course using instructional sequences, use interactive video to teach, or align content to educational standards as needed.

The point is this is very broad in its offerings, so how you use it as a teacher is up to you. You may wish to dive into the AI Tutor, which allows you to offer students real-time feedback on multimodal learning. Teachers can program this in a way that allows them to save time on individual support, while students still get that level of assistance while learning.

What are the best Kira Learning features?

Kira Learning is a powerful and rich offering so it can deliver an AI tool to work with most needs for teachers. For example, that means saving on time while still offering students personalized guidance and feedback, that can be monitored and analyzed by teachers to track progress.

Usefully, the Kira AI can adapt to student needs, so it will suggest targeted practice, offer useful insights to educators, and adapt explanations to help students as required.

This being education-specific, you can enjoy all the benefits of AI without any of the worries since this is compliant with COPPA and FERPA, uses AWS servers in the U.S., and is Common Sense privacy-verified.

The adaptive nature of Kira is really helpful. If a student keeps making the same or a similar mistake, for example, Kira will spot that and then adjust guidance accordingly. This should allow a new approach through which students find a way to understand and progress.

How much does Kira Learning cost?

Kira Learning is free for teachers and students, however, districts are charged at a bespoke rate for analytics.

Kira Learning best tips and tricks

Discuss first
Initially explain how this AI works so students are aware of its limitations and abilities -- and that they can ask you directly if needed.

Check in
The more you use AI, the more you can and should check-in to make sure it's working effectively and helping students as expected.

Teach CS
This is built from a computer science teaching platform so using it for that specifically is a great way to maximize what you're getting out of Kira.

• New Teacher Starter Kit
• Best Tools for Teachers

All these services integrate the science of reading best practices. Amira can also adhere to each district’s chosen reading curriculum.

Here’s a quick look at everything you need to know about Amira Learning.

What Is Amira Learning?

Amira Learning’s system is built upon research led by Jack Mostow, a professor at Carnegie Mellon who helped pioneer AI literacy education. Amira uses Claude AI to power its AI features, but these features are different than many other AI tools on the market. Instead of focusing on chat and generative response, Amira’s key feature is its advanced speech recognition and natural language processing capabilities, which allow the app to “hear” when a student is struggling and tailor suggestions to that student’s particular mistakes.

Though it’s not meant to replace a teacher, Amira provides real-time feedback and also helps teachers pinpoint where a student is struggling. For these reasons, Amira Learning is a favorite of education scientists and advocates for science of reading-based literacy instruction. The tool currently is used by more than 4 million students worldwide and across the U.S.

What Does The Research Say About Amira?

In one study by Columbia University, Amira's reading tutor demonstrated the ability to match human tutoring outcomes after just 30 sessions. Carnegie Mellon University researchers found that Amira’s read-out-loud tutoring approach significantly improved vocabulary in students.

These are just a few of many studies touting the benefits of the Amira app. The company notes that other studies have shown Amira outperforming human students at scale.

Amira is also constantly improving as AI continues to evolve. Amira is working with Digital Promise on a nearly $10 million grant-funded study to boost AI’s ability to recognize children’s speech overall, particularly children who speak English as a second language

How Much Does Amira Cost?

Amira is available through district-wide licensing, with flexible options specific to meet the scale and needs of each implementation.

Amira Learning Tips and Tricks

Check With Other Districts That Have Used Amira

Amira is used by schools in all 50 states, so if the tool is new to you, reach out to a district that has used it nearby and to their results. Utah has published a report on Amira’s efficacy in schools in the state, and in 2023 and 2014, the published data demonstrates that students who use Amira made significant gains.

Share Data With Students and Parents

A tool such as Amira is only effective if students use it. Reminding students and their parents that students who read with Amira outperform those who don’t can help encourage them to take advantage of the benefits Amira offers.

Read Aloud With Amira

One of the aspects that makes Amira unique is its ability to listen to students as they read. This feature is one of the best examples of AI edtech around, say some researchers.

Let Amira Help You As a Teacher

Using Amira as an AI assistant that can help coach students and identify areas they need to improve on can help make a literacy class more effective.

• AI Reading Tutors Are Already in Schools. Digital Promise Wants To Make Them Better
• The Science of Reading: What Teachers Need to Know

“AI has been here—it’s just forward-facing now and it’s here to stay,” says Bedden, who is wrapping up a five-year superintendency at Centennial School District in Pennsylvania. “There is uncertainty and a lack of awareness, and trepidation based on fear. But, for lack of a better term, it’s not the boogeyman.”

At the Networking Lunch and Panel Discussion at Tech & Learning’s EdExec Summit in Atlanta on August 19, Bedden will share the impacts of going big on strategic AI implementation in Centennial schools.

“We’re doing students a disservice if we don’t require [AI], and include work about responsible use,” says Bedden, who will be more than a month into his new role leading Village Charter School in New Jersey by conference week.

Bedden notes that before long, AI literacy will be as fundamental as traditional literacy. Such a statement “is absolutely controversial, but it’s our prediction,” he says. At college and in the workforce, AI will be part of students’ world, so talking through how to prepare students is educators’ responsibility.

Centennial’s AI implementation, launched in 2023, reflects leaders’ recognition that it’s far more than just another tech trend. Priorities include:

• Equipping educators and staff to safely and effectively leverage AI for innovative teaching, personalized learning, and efficiency.
• Establishing clear parameters and expectations for student use on generative AI in assignments.
• Building a foundation of responsible AI use that emphasizes transparency, accountability, and safety (including a mandatory district-developed AI Ethics Course for students in grades 7 to 12).
• Assessing AI tools continuously for age-appropriateness, potential bias, privacy protections, accessibility, and data security.
• Adhering to regulations (FERPA, IDEA, ADA, CIPA, COPPA) and district policies to ensure a safe, legally sound environment.

Centennial served as a model for this work when hosting the recent 2025 AI Symposium for Philadelphia region education leaders. In addition, through his consulting work Bedden is promoting how AI can create a more efficient, effective, and supportive learning environment, plus help schools reduce operational costs by optimizing budgets and minimizing waste while also meeting regulatory compliance requirements.

AI Efficiencies for Leaders and Staff

Technology implementations saving resources sounds like a budgetary win, yet one that could cost jobs. At Centennial, Bedden says, “We deployed without the intent to replace staff. If anything, it helped our staff be more efficient and effective, and in some cases we haven’t had to add staff.”

In his own work, using AI has been invaluable—resulting in colleagues being “in awe of how fast I’ve been able to deliver a response or product that has been requested,” Bedden says.

For example, recently he had to update the board on a school incident and got an assist from AI, with goals and guardrails. He asked for a summary of the incident and an email response that doesn’t violate privacy. “It gave me a succinct response and removed student names for me,” he shares. Then he asked for wording to show how the incident response adhered to a specific board policy.

When a colleague asks for assistance with a project, Bedden might share various AI-produced actions to consider, plus the iterative process he used for fine tuning the ideas. It doesn’t take long and he can quickly continue with his own to-dos.

Parents and other constituents interacting with Centennial’s website can now use an AI chatbot to get questions answered. “We call that a closed AI tool because it only pulls from our website,” Bedden says. “We didn’t want it pulling from outside. But if it doesn’t exist on the website, then people can’t give us feedback. It’s helping us improve transparency and service.”

AI for Teaching and Learning

Teachers and parents have had mixed reactions to the AI focus. “One thing we did from the get-go was communicate to parents, trying to get everyone on the same playing field,” Bedden says. “We wrote newsletters explaining what AI was and that we were going to be strategic. … Some people were hungry and excited about it, and a lot of people were like, ‘No, kids are going to cheat.’”

A video featuring three students sharing they already use AI—such as to get an explanation of a topic that’s different from a confusing one presented in class—helped show the need for educators to learn more.

Employee surveys made clear the need to address the technology’s human aspect. “You’ve got to be mindful of the false information, what they call hallucinations. They do exist,” Bedden says. “I always ask for evidence on a statement.”

Teachers get the final say on whether students use AI or not, and any cheating is handled like plagiarism. In their ethics course, students learn about properly referencing AI use and to “never look at it as the final say,” he explains. “As long as we do that, then we own the outcome.”

Bedden’s hope for AI is that educators accept it as part of our world. “But we also embrace our responsibility of keeping the human in the loop, and making sure users use it with a code of ethics,” he says.

• EdExec Summit

Spoiler alert: It won’t. At least not in the way some evangelists predict.

Why Tech Alone Doesn’t Transform Schools

Public education is not just a system. It’s a structure governed by high-stakes testing mandates, compliance-focused policies, union contracts, and deeply embedded cultural norms.

The issue isn’t a lack of innovation. It’s that the system is designed to resist meaningful change. It follows the money and prioritizes producing a compliant workforce.

We’ve Seen Better Models and Ignored Them

For decades, innovative school models that have broken the mold and defied conventional structures have emerged. These have proven that when you reimagine time, space, and curriculum, you can create places where learning feels real, relevant, and even joyful.

But today, most public schools have doubled down on test prep, pacing calendars, compliance, and seat time.

We already know what works. The real question is: Why don’t we adopt it?

What Other Countries Get Right About Innovation

Some nations do fund innovation, not just in theory, but in practice.

• In Israel, progressive schools, such as Democratic, Montessori, and Waldorf, are publicly funded on a per-pupil basis.
• In the Netherlands, the constitution guarantees freedom of education, meaning anyone can establish a school aligned with a specific philosophy (such as Dalton, Waldorf, or a religious one) and receive equal government funding.

In the U.S today., schools such as these are usually private, reserved for families with financial means. Public systems remain tethered to uniformity and test score data that primarily reflects socioeconomic status, English proficiency, and disability labels, making it a poor measure of factors under a school’s control.

If we can’t adopt successful, proven, human-centered approaches that already exist, why would we assume generative AI will force the change?

AI Can’t Fix What Culture Won’t Change

AI may auto-grade, differentiate, tutor, and respond instantly, but it won’t transform how we “do school” unless leadership, policy, and culture evolve.

Consider this: something as simple as the way we move students from class to class, by the ringing of bells, has remained unchanged for generations in most public schools. Students usually sit in rows, listen to lectures, memorize facts, and regurgitate information for tests.

And when AI emerged, what was the reaction of teachers, schools, and systems? Instead of integrating technology thoughtfully, many reverted to banning it, requiring students to rely solely on pen and paper. This approach, while well-intentioned, overlooks the fact that not all learners thrive with pen and paper, and technology can be an essential tool for many to express their thoughts and learn effectively.

If we haven't rethought these outdated fundamentals, why would we assume AI will suddenly spark a revolution?

Here’s What Will Revolutionize Education

If we truly want to revolutionize education, it won’t come from technology or AI. It will come from humans who embrace:

• Leadership that is bold enough to challenge the status quo and outdated norms
• Policy that funds and protects innovation in public schools, not just private ones
• Culture that understands and promotes student agency, relevance, and purpose

Until we empower public schools to implement what’s currently only available to the privileged, nothing will change.

The Revolution Comes from Us!

The revolution doesn’t come from smarter machines. It comes from smarter humans—leaders who dare to change policy and foster a culture that will open doors to innovation for all. That revolution won’t come from AI. It will come from us.

• Navigate AI with New Risk Assessments from Common Sense Media
• Lights, Camera, and 7 Actionable Ways to Use Google Vids

Of course, as you’ve already guessed, even though the people in the video look, sound, and appear to be real, they were indeed created with “prompts” by AI — specifically Google’s new VEO 3, the tech giant’s most advanced public-facing AI video model.

VEO 3 is currently available to subscribers of Google’s $250 per month AI Ultra plan, which makes it unavailable for many education budgets. Yet, it’s still technology that seems likely to influence education in positive and not-so-positive ways.

On the positive side, once the technology trickles down and becomes more available, it’s easy to imagine educational-use cases with tons of options to create videos that relate to class exercises in various subjects, such as history or STEM.

But this technology also highlights the gray area around AI use and what its impact on actual creatives might be. Beyond that, it will -- troublingly -- lend itself very well to deepfakes. AI videos have already been used to spread misinformation, and to bully and harass in school settings. A tool such as VEO 3 potentially makes those types of illegitimate, and in some cases criminal, uses easier to create and even harder to spot.

Here’s a closer look at everything educators need to know about VEO 3.

What Is VEO 3 and Why Is It Different Than Other Video Generators?

VEO 3 is Google’s latest and greatest AI video generator. Combining Google’s Veo, Imagen, and Gemini AI models, VEO 3 is designed to let users create short, life-like videos with audio, and then string these videos together in longer scenes.

The tool is designed to make writing prompts for these videos easy and intuitive. This audio and ease of prompting in particular stand out when compared to other Google video tools and the competition. For example, when I used OpenAI’s Sora video generator, there was no audio, and prompting was as unpleasant as high school calculus.

But reviewers say VEO 3 not only produces amazing clips with audio, it does so with relative ease, though VEO 3 is far from perfect.

What Are Some Ways VEO 3 Can Be Used For Teaching?

If VEO 3 becomes more widely available for less money, or if your school has a large AI budget, it doesn’t take much to come up with creative uses for it: Writers can bring their stories to life with prompts, history students can recreate scenes from events they’re studying, and biology students can model photosynthesis and assess the video’s accuracy.

VEO 3 also has a feature that lets you see exactly what prompts others used to create their videos. For the growing field of AI education and prompt writing, this is a valuable feature and one that educators might encourage their students to explore in the future.

What Are Some Concerns About VEO 3 Technology?

VEO 3 does not yet have a tool that lets users upload an image and generate a video based on that, but that tool is available for VEO 2 and is expected to be rolled out to VEO 3 eventually. This will make VEO 3 a particularly powerful tool for generating deepfakes.

Even without this, VEO 3’s realistic videos and ease of use make it likely to increase the use of AI to spread misinformation and potentially bully and harass people. This is true for society in general, but will be amplified in school settings.

What Else Should Teachers Know?

Regardless of whether you’re a fan of AI technology or not, learning about the capabilities of VEO 3 and other AI video generators is critical. These tools are here, for better or worse, and educators and their students need AI literacy training in order to navigate a world in which what we see is not always real.

We’re not all prompts, as the video referenced at the start of this article claims, but increasingly much of what we read and see on the internet will be created from prompts, and we all should be thinking about how to navigate that.

• Sora: Is OpenAI's Video Generating Tool Helpful For Teaching?
• Lights, Camera, and 7 Actionable Ways to Use Google Vids

Teachers may not need to imagine too much longer. AI already provides helpful reading tutors and could have all the capabilities described above in the coming years, says Jeremy Roschelle, Digital Promise’s Director of Learning Sciences Research.

Digital Promise received a nearly $10 million grant from the Institute of Education Sciences to launch and lead the Using Generative Artificial Intelligence for Reading R&D Center (U-GAIN Reading). U-GAIN Reading will build on existing research conducted by Amira Learning, a research-backed Intelligent Tutoring System (ITS) already being used by more than 1 million students each year.

The goal of U-GAIN is to turn the page on many of the literacy struggles that students across the nation struggle with by helping to provide research-backed evidence and training data to improve the ability of AI tutors to recognize diverse student voices and further engage students.

“Learning to read remains a national challenge with declining NAEP scores, and those scores show that many students are struggling to read even at a basic level by fourth grade,” Roschelle says.

As any teacher knows, these types of struggles often have heartbreaking consequences.

“If you learn to read, you learn every subject better, you end up with better outcomes in terms of going to college, better finances, better health," Roschelle says. "Learning to read is just such an important predictor of a student's further trajectory.”

AI and Literacy: Hearing Student Voices and Boosting Engagement

AI reading tutors already have impressive abilities. Amira, for instance, can listen to students read and offer feedback and assistance on mispronounced words and more, but the U-GAIN team hopes to take these capabilities to the next level when it comes to understanding what students are trying to say and recognizing their engagement levels.

Listening To Students

“For this thing to work, it has to accurately hear kids,” Roschelle says. “Many students speak with a dialect or an accent, or from a different regional idiom, and when this kind of technology can't hear a kid accurately, it can't help them.”

He adds, “Teachers experience their kids on speech recognition products, and they tell us the pain [they experience] when a multilingual learner, an English learner, speaks and the teacher can understand what the student is saying, but the technology can't.”

This problem arises because most speech recognition AI doesn’t have enough training listening to kids speak overall, and kids with various accents, in particular.

To help change this, the U-Gain Reading team is hoping to listen to between 500 and 1,000 kids who are using Amira weekly for a half year or more. They are working with linguistic experts from multiple school districts, and are optimistic they can get appropriate permission forms from enough parents to create a robust new dataset on which to train the AI tutors.

“We'll be able to make a big improvement in hearing kids speak,” Roschelle says.

AI Engagement

A student can sit with an AI reading tutor for half an hour each week, but if they’re not engaged, they’re not going to get the most out of it.

“We need that half hour per week, minimum, but they need to be really cognitively engaged in reading, which isn't always fun when you're struggling to read,” Roschelle says. “The really big challenge is many of the techniques we have for detecting engagement works with clicks and work with keystrokes.”

To be effective in this setting, Roschelle says, we need tools that can measure engagement through thing such as the tone of a child's voice in the same way a teacher might be able to.

Harnessing AI's Potential For Individualized Learning

After challenges around engagement and speech recognition are overcome, AI tutors can respond to students in new and robust ways.

“Once we have those, then we have these amazing capabilities of generative AI that we can start to apply to solve problems,” Roschelle says. In particular, Roschelle is excited about the possibility of generating unique readings and responses for each student.

“Most products, Amira included, have a corpus of texts that they can give to a student to read, but they don't necessarily engage every student’s background knowledge,” Roschelle says. “And you read better — this is pure science of reading— when you engage background knowledge and you build background knowledge. Reading isn't just decoding. It's also using knowledge.”

An AI that can build higher-quality customized texts can potentially boost student engagement with the lesson. “We know if kids like dinosaurs, give them more dinosaurs,” Roschelle says.

AI tutors also need to learn to better respond verbally to students who are struggling with a text. Since many new readers are by definition unable to read, responding in a conversational way is key.

“You have to give them verbal feedback,” Roschelle says. “Right now most of the products do that in a limited way with canned interventions.”

He adds, “What we're on the verge of is being able to generate conversations much more like what a parent or a teacher would have with the kid that feature both the engagement and the reading skills, side by side.”

Roschelle is excited about what the U-GAIN Reading team will learn over the next five years, and believes with the help of a skilled teacher, AI can start to make more of a dent in some of the reading struggles students currently have.

“We’re working on the tough problems that will really drive the field to the next generation,” he says.

• AI In Education Executive Order: What You Need To Know
• 4 Studies About AI Tutors Every Teacher Should Know

The panel featured:

• Matthew Winters, AI Education Specialist, Utah State Board of Education
• Dr. Michael Gaskell, School Principal in East Brunswick, New Jersey, Author, and Speaker
• Kathi Kersznowski, Lead Educational Technology Specialist, Washington Township Public Schools in New Jersey
• Kellie Ady, Senior Director, Education Strategy, PowerSchool & Government Relations

The participants shared advice, best practices and specific ways that they personally approach providing PD centered around implementing AI into their work.

Watch the on-demand webinar here

Key Takeaways

More Engaging AI PD - “I love to make PD fun, and people don't wanna work in the summer,” said Kersznowski. “So if you can make learning not feel like work, I think that's your best bet." She shared suggestions to make AI PD more engaging and informal, including:

• AI Appetizers: Friends gather, bring appetizers, and informally explore AI tools such as ChatGPT or Claude.
• AI Accountability Buddies: A group text in which colleagues share AI discoveries and ideas.
• AI Cook-off: Participants challenge one other to use AI to generate recipes and then prepare them.
• Daily Coffee Chat: Anyone can spend five minutes each morning trying out an AI app on their phone.
• Spontaneous Planning: Use AI to generate fun day trip ideas based on preferences and budget.
• Playing with a Podcast (NotebookLM): Using NotebookLM (a free Google product) to create conversational podcasts from prompts, topics, websites, or YouTube videos. This can be used for personal interests or to spark ideas for classroom use.
• AI Book Club Member: Adding a chatbot as a member of a book club to offer its thoughts on the reading.

Find game changers - Gaskell suggested keeping it simple, focusing on “The whole concept of, ‘Wow, that's something fairly simple, straightforward, and very intuitive that that teachers can get interested in,’” he said. He highlighted "game changers" that can excite educators about AI, focusing on practical and intuitive applications. He then shared personal examples of using Google Gemini to:

• Convert a handwritten list of calendar dates from a Post-it note into Google Calendar events
• Create a calendar event from a screenshot of a confirmation text message (such as for a service appointment)

He also advocated for using NotebookLM for PD, noting its ability to create conversational voices and the new beta mode allowing interaction with them.

Start by getting personal - “I think everyone's aware that we have sort of a gap between student adoption of AI tools and then teacher adoption of it at the classroom level,” said Ady. “But we are seeing some shifts in terms of being more open to trying AI tools instructionally and not just personally.”

To help teachers who are reluctant to adopt AI tools, she suggested:

• Remind them about the tools they already use in their personal lives, such as GPS.
• Encourage using AI as a "thought partner," whether with a chatbot or a colleague.
• Highlight how AI can save time, such as by communicating with families.
• Embed AI tool usage directly into PD, allowing them to experience AI as learners.

Encourage play - "What we're learning is the importance of play,” said Winters. “I think we, as teachers, don't do enough of it in our classrooms, but also we don't play enough as adults. And when we talk about a tool that rewards creativity and curiosity like AI does, especially generative AI, we really gotta jump in and play once in a while with it."

Winters also discussed Utah's statewide AI training initiative, and emphasized how AI can boost productivity for teachers by:

• Speeding up actions: Using AI for tasks such as writing emails, creating formative assessments, and drafting rubrics, with human editing and revision.
• Fostering creativity: Exploring new possibilities within the classroom system that were previously unknown.
• Enhancing accessibility: Supporting traditionally underserved populations, such as students with disabilities, multilingual learners, and gifted and talented students.

He also suggested focusing on "tool types" rather than specific AI tools due to varied district resources and safety guidelines. He categorizes these as:

• Constitutionalized chatbots: AI with added rules for student functionality.
• RAG (Retrieval Augmented Generative) databases: Tools that draw from a knowledge base.
• Multimodal AI: Tools that produce various content types such as images and text, or even videos.

Looking ahead - The participants were asked for a bold prediction about AI.

"I don't know if this is a bold prediction, but I think it's a must do,” said Kersznowski. “AI is here. I think if it didn't settle in last year, this past year, I think everybody has started to realize it's here to stay, and it's not going away, and it's nothing we can continue to fear."

Winters added: “Ken Robinson, one of my favorite educational theorists, says that PD should be viewed not as an expense, but as an investment. And I think when we talk about artificial intelligence or anything that's associated with personalized learning, we need professional development consistently for teachers across the country to help build that capacity and around the globe as well.”

• Tech & Learning webinars

In practice, Arc works like a beefed-up, more accurate, and more user-friendly version of Google’s AI summaries. Search for a person’s name, and Arc will quickly create an engaging dossier of results, drawing from their LinkedIn profiles and many other sites to give you an instant, and in my experience, accurate snapshot.

You can also type in academic queries such as “How to write a research paper” or “How to solve for x,” and Arc will summarize the main advice on these topics and provide links. If you’re using the Arc app, you can pinch on specific links, including video links, and get an instant summary of their content.

I started using Arc over the last couple of weeks and have quickly become addicted. It has a fast and visually appealing interface that helps me find what I’m looking for more quickly, though it does have some minor drawbacks. Its summaries sometimes emphasize the wrong points, and I worry about what this type of technology will do to the content creators whose work Arc and other similar tools are mining.

Overall, however, in terms of user experience alone, Arc is a winner, and there are many educational applications for its use.

What is Arc?

At its heart, Arc is an AI-powered search engine. It summarizes and offers customizable features to let users reconfigure what browsing the web looks and feels like. That seems like a bold goal, but one that in my mind Arc has largely accomplished.

I became aware of the app recently when a person I was interviewed alongside shared that they had used it to look me up before agreeing to the interview. Curious, I searched my name and found that it quickly pulled from my existing bios on this and other websites, as well as sources such as my LinkedIn profile, to create a quick snapshot of who I am as an educator and writer.

I was impressed with the accuracy and extent of the overview. You could easily get all the information provided without this tool, but you’d have to click two or three times instead of just once. You might also be drawn to a specific area of my work and be given a less holistic view.

What Does Arc Do Well?

Arc excels at providing a quick and general overview of whatever you search.

For example, many of the undergraduate students I am working with are struggling with proper APA formatting, particularly citations. I searched “advice for APA citations” on Arc and quickly got an overview that I will share with my students, as well as links to resources I can share with them as well. The way Arc laid out this info was clearer, quicker, and more user-friendly than resources I’ve found using Google in the past.

Similar results occurred with other topics I searched. I’m finding Arc is a good tool for helping me find resources to direct my students to right now, and could potentially be a helpful tool for students themselves to use in the future if it was approved for student use at the institutions at which I teach.

From a design perspective, Arc just functions really well and smoothly. Some of the best and most powerful AI tools, including those from OpenAI and Google, still suffer from an often slow and almost always clunky interface that can detract from the time-saving goal of turning to AI in the first place. That doesn’t happen here.

How Could Arc Improve?

A good way to test Arc’s accuracy is to search your own name. When I did this, I was impressed overall but thought it emphasized some strange things. The first university it mentioned me teaching at is one I have not taught at for half a decade. It also identified me as living in Danbury, Connecticut. I don’t live there, and never have, though I live nearby.

These kind of little mistakes could leave you or your students with some pretty bad conceptions depending on a topic, so as with all AI, it’s not a source that should be trusted without verification.

Beyond the app’s functionality, I worry about what this type of technology will do to content creation. I’m not a huge fan of the click economy and believe that chasing views has been the original sin of the journalism industry in my lifetime.

I’m also a realist, however, and know that many jobs, including many traditional journalism jobs, are paid for by ad revenue generated by clicks. This type of technology, and Google’s similar efforts to update searches, have the potential to obliterate the modern internet economy. This might have more repercussions than is apparent at first glance. The APA citation advice Arc provided, which I praised previously in this story, drew from websites and YouTube videos made by content creators, who often rely on views for their income. But I didn’t view their work before using the advice. Eventually, this could disincentivize many forms of online content creation, from writing to videos to even podcasting.

Even if you don’t care about the content creators themselves, you need to remember that they are the source of the AI’s knowledge. As a recent Atlantic article put it: Google’s New Search Tool Could Eat The Internet Alive.

How Much Does Arc Cost?

Arc’s browser and app are free, and the company says they will always offer free versions, though they may have some paid features in the future.

Bottom Line: Is Arc Worth it For Educators?

If you’re looking to explore different AI use cases and/or just want a quicker and potentially more efficient way to surf the web, I recommend exploring Arc. Despite some misgivings about the long-term impact of this type of technology, I find it helpful and love its smooth interface. Even the AI-powered versions of search offered by Google and Microsoft feel old-fashioned in comparison.

• Recognizing When Not To Use AI
• Optimizing Teacher PD with AI

What do in-person AI labs look like and how are they being used? Following is a glimpse at what the AI action is like at four institutions: Pace University, Foothill College, Sacred Heart University, and Bryant University.

How Do Institutions Describe Their AI Lab?

AI lab descriptions can lean explanatory or conversational.

Pace (New York City) and Bryant (R.I.) take the explanatory route:

• Bryant's AI lab main page describes the lab as “a hub for students and faculty to collaborate and explore the potential of AI in business. This dynamic space serves as a showcase for Bryant’s data-driven and innovative culture and promotes the Data Science and Information Systems programs to prospective students and corporate partners.”
• The top of Pace’s page notes, “Students, faculty, and staff come together to research and learn about artificial intelligence—from machine learning and natural language processing to powerful prompts for generative AI.”

Foothill College (of the Foothill-De Anza Community College District in Calif.) and Sacred Heart (Conn.) use the more conversational approach:

• “Welcome to the Foothill College Artificial Intelligence (AI) Lab – The AI lab is where we explore and learn about Artificial Intelligence, Machine Learning, and Robotics.”
• “Immerse yourself in the latest AI technology and the chance to collaborate on useful, new applications in our state-of-the-art facility,” beckons Sacred Heart’s AI Lab page. The lab “offers the cutting-edge tools needed to tackle new and arising challenges in our modern world.”

What Needs Do AI Labs Meet?

Pace’s lab, according to an associate dean in a March 2024 announcement, allows the university to “consolidate, strengthen and extend the existing AI initiatives and create exciting new ones in education and research-leveraging partnerships within and outside the university.”

Bryant’s “why” involves allowing students and faculty to collaborate and explore what AI can do in business, plus showcasing how the university has a data-driven culture as well as promoting its data science and information systems programs to prospective students and corporate partners.

Sacred Heart describes its lab as a collaboration to address issues in the world and “collectively drive the contemporary run into digital revolution.” In the process, students learn “how to manage a project and bring a project to completion,” says Bob McCloud, a computer science professor. “In AI, the most important thing we teach them is, what is good data, what is reliable data, and how to structure data so you can use in AI process.”

Foothill, meanwhile, is one of dozens of community colleges collaborating with the American Association of Community Colleges, Dell Technologies and Intel to grow the AI Incubator Network and Intel’s AI for Workforce programs.

Who Can Use AI Labs?

These labs are typically billed as open to all faculty, students and staff. Pace calls it “the central hub of AI for the Pace community, offering training and opportunities to learn, grow and collaborate.”

Internal (students in related courses, the esports club team at Sacred Heart, the student-run Google Developer Group at Pace, for example) and external (like the New York tech community at Pace, invited to network with students) groups are more targeted AI lab users.

What Equipment Do AI Labs Have?

Students and other visitors at Pace bring laptops and connect to the high-performance computing (HPC) and Nvidia GPUs in the lab, says spokesperson Sean Coughlin. Additionally, three Alienware workstations are available. In the same building, the separate Robotics Lab offers hands-on training using a Clearpath Humanoid robot, a mobile manipulator and Clearpath Jackal devices used in computer vision education and research. Plus, Coughlin says, “the lab sports a Shadow Robot Hand.” Ph.D. students conduct AI research, using the robots for experiments and to develop models.

Bryant’s AI lab has a dedicated space for robots and includes two humanoids (Pepper and NAOv6), two vehicle robots, and robotic arms. It houses five collaborative GPU workstations with chips manufactured by NVIDIA, plus movable whiteboards (for more workspace and as privacy partitions) and large-screen monitors. “The AI Lab is designed for students to experiment with machine learning and its applications in robotics,” says Chen Zhang, an associate professor of information system analytics.

Forty Alienware computers with advanced processing capabilities and 24 separate workstations that can access three dedicated servers for student and faculty research are found in Sacred Heart’s lab, which also has object recognition equipment and eye trackers.

What Are Some Ways AI Labs Are Used?

Foothill’s AI lab has supported the college’s Tools for Transition and Work program supporting neurodivergent individuals. Four students in a winter 2023 pilot program spent their academic quarter building autonomous robots and learning engineering and coding skills. The number of participants has since grown to at least nine. Meanwhile, staff and faculty provide AI exhibits and demos at college events, including those for youth from local organizations.

Sacred Heart’s lab has helped the local community, too. In one project, professors and students studied how to improve a municipal tree planting program between the town and homeowners, explains McCloud. The study identified the best tree types and providers for various circumstances.

In another project, Sacred Heart students, professors, athletes and coaches looked at the correlation between time played and performance. The work involved monitoring training, statistics, sleep minutes and game minutes to predict optimal training levels.

“It seems as if some players are playing too many minutes for optimum performance,” says McCloud. “The idea of AI is to be able to look at reliable data and learn from it and then keep learning.”

And in an ongoing study, convolutional neural networks are being used to show how fraud can be detected in painting attribution.

Pace students’ experiential opportunities have included the AI Internship Experience Program. Using the lab, student teams developed AI models that could classify and generate images of butterflies, flowers and Minecraft-inspired visuals. Also, faculty developed a module for a Design Faculty Global Network course, covering equity-centered AI design. About 15 Pace students took the online course, taught by professors from Columbia, Finland, Portugal and the U.S.

Bryant data analytics students have used the AI lab to research a project involving AAA travel booking data and how it can predict whether and why some people cancel their trips. Corporate partners can use the AI lab for meetings, presentations and research, with the hope that students with expertise in AI, machine learning, analytics and robotics could participate, plus serve as a recruitment pool.

When students have opportunities to work with AI, they’re not only building technical skills but also becoming skilled in project management and bringing new information to the world.

“Students learn how to define scope, they learn how to say what they’re going to do in a project at the beginning,” says McCloud. “We focus on constructing or importing accurate, reliable data sets and then figuring out how to use those data sets to increase knowledge.”

The company says its free resources are used by 5 million students every month. It offers a wealth of information, including slides for teachers and how-to guides for students.

This covers a range of services from proofreading and editing to plagiarism checks and citation generation. The idea being that students should be able to get to a top-grade paper by enhancing their efforts using this service.

This guide aims to explain all you need to know about Scribbr in your class.

What is Scribbr?

Scribbr is a paper-writing aid, backed by real people as well as AI resources, that aims to help students write the best possible papers they can manage. It also represents a helpful resources for teachers to use when grading, and to help students write their papers.

The service is built to cover a few main services, which include: proofreading and editing, plagiarism and AI content checking, citation generation, paraphrasing, and more.

It offers more than 300 helpful articles, at time of publishing, that touch on topics that can assist students, including academic writing, plagiarism, citing sources, and more.

The system works with a variety of papers, from essays and theses to PhD dissertations and research proposals. The idea being that this should help get it perfect before handing in to be read and assessed.

How does Scribbr work?

Scribbr comprises several different tools with the human-powered ones in the more premium offering as well as some more AI-powered services in the free tools. If you opt for the paid submission, then your paper can receive the human checks, which generally get turned around in a 24-hour period, although it can take as long as a week, depending on how long your paper is.

While the human checks are well-suited to proofreading, editing, spelling, grammar, and citations, the AI tools can work well for these, too. Although its the citation generator that works very well. This allows you to have your essay formatted with citations in a few clicks, and even lets you select the output style, including MLA, APA, Chicago, Harvard, and others.

The plagiarism detector is a useful tool for teachers as it uses advanced dynamic databases to work out any attempts at plagiarism as well as AI detection being possible.

What are the best Scribbr features?

Scribbr uses a plaigarism checker that's powered by Turnitin as a way to offer excellent accuracy. Thanks to this being used in multiple libraries and databases, it has wide-reaching access to materials so the results are generally very accurate and reliable.

The use of AI tools also allows for use by those who aren't able to pay for the human touches, or are limited on time. These can also be a useful second line of defense, perhaps running a paper through these first before sending in for a human edit to ensure total perfection -- or as close as you can get in real life.

The multilingual offerings of this service are also worth noting as this might be primarily aimed at English writers and speakers but goes beyond. You can take a look at the many editors working with Scribbr to see the varied language speakers available.

Customer support is also excellent, which can be seen from the Trustpilot page alone, where all concerns are addressed and all issues are mended by the attentive team.

The Summarizer tool is really helpful when researching as it can make ingestion of large and complex bodies of information far quicker and more easy.

How much does Scribbr cost?

Scribbr offers a selection of free resources as well as its paid-for premium services that are charged for depending on the paper.

The rates for papers are $19.95 for a small document up to 7,500 words, $29.95 for a medium document up to 50,000 words, and $39.95 for a large document of more than 50,000 words.

Scribbr best tips and tricks

Start with AI
Use the free tools first including AI and then submit if you want a two-layer approach to catch any possible mistakes.

Mark with AI checks
Use the AI checker to ensure students aren't cheating and show them how it works so they realize AI can't be used as a shortcut.

Use the Summarizer
Have the Summarizer tool condense papers when researching to cover more ground if working through lots of learning material.

• New Teacher Starter Kit
• Best Tools for Teachers

I recently tried using Scite as an AI-powered research assistant for the work I do both as an educator and journalist.

Overall, I’m impressed with Scite’s ability to quickly find and summarize research papers from a diverse range of academic disciplines. It does this quicker than other AI research tools I’ve used, and has an easy-to-use interface. The only downside is that it requires a $12 per month subscription, and like all AI tools, won’t immediately find all the best sources on a given topic.

Therefore, whether you’re a student or academic using this, you’ll likely still have to do additional digging and searching for sources the old-fashioned way. This isn’t always a bad thing. In fact, one of the things I love about Scite is it encourages you through the research process by eliminating some of the barriers or difficulties around getting started. In other words, it is the type of AI tool that me and many other educators are always talking about wanting: one that facilitates rather than replaces thinking.

Here’s a closer look at everything Scite does well and where there is room for improvement.

What Is Scite and How Does It Work?

Scite works like a beefed-up AI-powered academic search engine.

I asked it “Can you summarize research on the use of AI chatbots as tutors in education?” With a minute or two, it provided an accurate overview of the current research into this topic with citations linking to the latest studies.

The summary noted that “AI chatbots are increasingly recognized for their ability to engage students in a personalized manner,” and shared the researchers making this claim and their evidence for it. Then it summarized some of the concerns researchers have.

Scite noted that “Despite the promising advantages, there is also critical discourse surrounding the implementation of AI chatbots in education," and that some researchers "caution about the hype surrounding AI chatbots, suggesting that while they exhibit considerable potential, their actual effectiveness in replacing traditional pedagogical methods requires comprehensive evaluation.”

Scite also shows you how often a paper has been cited, and lets you quickly view this citation, so you can tell if it was cited in a positive or negative way.

What Does Scite Do Well?

The Scite features described in the previous section are a really welcome resource for someone conducting a deep dive into any academic topic. This quick snapshot of the overall consensus around a topic is a wonderful place to start research. I would definitely recommend it for this purpose to students and fellow instructors.

The ability to quickly assess how often the paper has been cited, and in what context, gives you an almost instant sense of how the paper is viewed by other experts in the same field, which is so important given how one paper can sometimes find results that are seemingly amazing but can be so out of whack with other findings on the topic that it strains credibility.

Of course, there are some caveats. Scite does not have access to every academic paper, so students, in particular, need to be reminded not to begin and end their research with Scite.

Scite is not the only helpful AI research tool available, but it provides a quicker and more concise overview of research than others I’ve used, including STORM and OpenAI’s deep research tools. Both STORM and OpenAI’s research tend to provide longer and more in-depth overviews about any topic and write something based upon that topic for you. This might be preferable in some settings, but for me, the shorter overviews Scite provides are more useful and do what I want: help me find relevant research while leaving the actual writing to me or my students.

What Are Scite’s Limitations

Scite is a great tool, but it should not be your only research tool.

On some topics I investigated, I found several important studies that Scite missed or deemphasized. The tool also has a tendency to move toward centrist views on every issue. This is probably for the best overall, but doing that potentially risks overlooking some important minority viewpoints. And all summaries, even human-written ones, by their nature can gloss over nuance, particularly in discussions of complex topics, which most academic inquiries tend to examine.

So if I were to get approval from my university’s security and IT team to recommend this tool, I’d definitely remind students of these concerns. I'd also stress that while Scite can be helpful to use, it doesn’t replace conducting additional research on their own.

How Much Does Scite Cost?

Scite costs $12 per month for individual users. Additionally, it offers institutional subscriptions for companies and universities, but those are custom and based upon the number of licenses needed and other factors.

Bottom Line: Is Scite a Good Tool For Teaching?

Overall, Scite is a helpful and engaging research tool that can likely benefit educators, academics, and students. It requires a monthly submission but seems worth it to those who regularly conduct research.

Ultimately, I found Scite to be one of the best AI research assistant tools around.

• Recognizing When Not To Use AI
• 10 Free AI Trainings and Certifications for Educators

Despite these efforts, inappropriate AI use has increased among my undergraduate students instead of decreasing. I’ve noticed what I believe is AI writing not just in formal assignments, but in discussion board posts and, perhaps most disturbing of all, in written responses to classmates and to me.

I’m not alone. Reliable data is still lacking on how prevalent AI writing is in education, but in casual conversations with colleagues, most have reported similar upticks, with several estimating that more than 50% of writing submissions are now AI-generated.

Recently, I attempted to fight fire with fire, and asked ChatGPT how I could prevent students from using it and other similar tools to write their work for them. ChatGPT responded with a range of suggestions. Some were standard but helpful, others were standard and I think ineffective, while there were also a few tips I haven’t considered implementing.

Here’s a closer look at ChatGPT’s advice for educators such as me who are, ironically, tired of reading and grading papers written by ChatGPT.

Asking ChatGPT How To Limit Student Cheating

For this experiment, I used GP4-4o model, which I get access to through my subscription to ChatGPT Plus.

I started with the following prompt:

Hi, I'm a college writing and English professor. My freshman writing students are frequently using ChatGPT and similar AI tools to write their papers and their online discussion board posts. What can I do to prevent them?

ChatGPT provided a series of suggestions. I then asked it to share an example of an AI-resistant assignment and look at one of my assignments, and followed up with other requests. The advice ChatGPT provided on preventing ChatGPT fell broadly into the following categories.

1. Make Prompts Personal and Specific

What ChatGPT advised: Ask students to connect readings or themes to personal experiences, class discussions, or campus/local events. Use hyper-local or course-specific references: “Refer to our in-class debate on March 5th” or “Discuss your experience attending the campus event on X.”

My take: This is really good advice. I teach online, so it’s not always applicable for me, but overall the principle is sound. Focus on assignments that require information that is not available on the web. This way, even an AI equipped with web-browsing capability can’t come up with satisfactory answers.

However, when ChatGPT provided examples of writing assignments it felt were AI resistant, ChatGPT didn’t quite follow its own advice. For instance, one assignment asked a student to include a personal reflection on what writing the assignment was like. In my experience, those types of questions are easy for AI to duplicate.

2. Require Process Documentation

What ChatGPT advised: Have students submit brainstorming notes, outlines, rough drafts, peer reviews, and reflections. Ask for screenshots of revision history if using Google Docs. Include a short oral presentation or in-class writing follow-up.

My take: This is sound advice. I’ve resisted implementing these types of documentation requirements in my own classes because it's a cumbersome process for me and feels like an annoying extra step for those students who are not using AI to do their work. However, it seems increasingly necessary, and if structured properly, can help reinforce lessons around outlining and the overall writing process.

3. Use Multimodal or Interactive Assignments

What ChatGPT advised: [Use] podcasts, infographics, video essays, or collaborative wikis to make AI generation harder and student voice more visible.

My take: This is great advice and can make assignments more interesting and fun for students in general. For writing class this is a bit more challenging because my job is to teach students to write, not necessarily to create podcasts or audio content. However, I'm always looking for new and engaging ways to convey writing concepts, and this could do that with the added bonus of limiting the impact of AI.

4. Integrate AI Use Transparently

What ChatGPT advised: Instead of banning AI entirely, consider teaching responsible use: Scaffold AI literacy. Give students assignments that require them to experiment with AI tools and reflect on their usefulness and limits. Example: “Use ChatGPT to generate a rough draft of an introduction, then revise it critically. Highlight your changes and explain your reasoning.”

My take: I like this advice in theory, and teaching AI literacy and appropriate AI use cases is something I support. However, in my experience, it doesn’t work quite as smoothly as this suggestion makes it seem. My AI-approved assignments are just as likely to be AI overused as non-AI-based assignments. Once students are using AI to generate a rough draft, many also use AI to generate their reflection on the assignment, in my experience.

In this vein, ChatGPT did suggest an interesting assignment in which you have students read AI-generated work and student-generated work and guess which was which. The hope is that they will see why the AI-generated work is lacking. I like this suggestion even though the "encourage specific types of AI use," has not been an effective strategy for me.

5. Use In-Class Writing to Anchor Authentic Work

What ChatGPT advised: Require timed, low-stakes in-class writing assignments or reflections. Use in-class work to build parts of a longer paper. This creates a clear benchmark for each student's natural voice and fluency.

My take: I like this advice. Though it is hard to implement in the online classes I currently teach, I used to devote a good deal of time to in-class writing when I taught in person for other reasons. Working with students as they write, rather than criticizing them afterward, is a good pedagogical policy overall, I think. Here, it has the added benefit of allowing you to help cut back on AI use.

Overall Takeaways

Like many instructors these days, I spend a lot of time grading ChatGPT-produced work supposedly from students. I generally have a lot of complaints with these assignments, but in this instance, I think ChatGPT earned an A.

The advice it provided was a quick and accurate summary of what most instructors who have thought about these issues would advise, and gave me some food for thought on how I can change my approach to decrease AI use. For example, I will likely spend more time focused on activities such as in-class or synchronous writing and documentation.

The best AI counterstrategy, however, seems to me to be hyperlocalized assignments that require offline reporting or research. If this can be realized, it appears to be the most direct way to beat the machines without making things harder for us instructors or unnecessarily laborious for students.

Even ChatGPT agrees that’s a win-win.

• Optimizing Teacher PD with AI
• Recognizing When Not To Use AI

“Agricultural education really was STEM before STEM was an ubiquitous icon for all things science, technology, engineering, and math,” Stump says. “FFA was formed in 1928, and all of it has been about really building the science, business, and technology skills within the next generation of those who are going to be producers, either in the form of a farmer or rancher, that are going to be the veterinarians, the chemists, and the food safety experts that are ensuring that this country can continue to provide the world's most abundant, safest, and most economical food supply across the globe.”

Stump, a former classroom agricultural educator, is passionate about how integrating farming education into existing STEM and CTE curricula can help build engaging lessons that not only help students understand the importance of STEM concepts and the real-world applications but also expose them to important career opportunities in the agricultural industry.

Farming-Related STEM and CTE Resources

FFA is a school-based leadership organization with chapters in all 50 states, Puerto Rico, and the U.S. Virgin Islands. The organization works with educators throughout the country, looking to incorporate farming and agricultural lessons, and provides additional tools for teachers via its website.

“We have a whole educator resources section,” Stump says. “If teachers want to take biology and really make it practical for students, if they want to take chemistry and make it practical, we have a wide variety of quick resources to kind of integrate AG as the application.”

He adds, “Over the past two years we released a full suite of resources on sustainability in agriculture that walks through all of the practices that farmers and ranchers and everybody across the food value chain are taking and implementing to ensure that we have the smallest carbon footprint that we can have and actually be a part of the solution.”

One example of this is how large-scale farming can sometimes create carbon sinks, a system in which carbon dioxide is removed from the atmosphere, Stump says. “The Midwest, with all the corn growing in the summer, actually pulls in a lot of carbon out of the atmosphere. How do we do that better?”

Answering this question, “Is one of the solutions we're continuing to work forward and connect students to,” he says.

Farming Education and Career Opportunities

Stump says some schools have large outdoor farms that act as living labs for students in various science classes. Even schools that don’t have those resources can still get students outside and working with plants.

Back when he was working as a teacher, Stump says he would take his students outside to the school farm with the biology teacher. “We would actually section off a certain area, and our job was to specify all of the life forms that were in that space, both plant and insect, and really try and identify what the healthiness of that ecosystem is,” he says. These types of activities truly become a way of bringing that learning to life in a way that sparks innovation, he adds.

Many students who move from these farming-centered classes go into providing innovation in the industry. For example, one recent FFA alum is developing a company that has designed an early warning system for blight by using a plant’s natural ability to change color when it becomes distressed, Stump explains.

The FFA maintains a resource of more than 300 career opportunities related to agriculture. These include science and biology roles as well as production roles and engineering and design roles related to equipment that supports the industry.

AI and Agriculture

Beyond chemistry and biology, data science and AI are also playing an increasing role in agriculture and agricultural education. The FFA is working with many schools to integrate precision farming into the curricula, which uses smart sensors, data science, and AI.

Recently, the organization partnered with Microsoft to bring FarmBeats, a precision agricultural program, to more than 185 middle and high schools. The program provides schools with a curriculum and kits that include a ready-to-use agricultural sensor system.

“Schools get basically a small motherboard that connects to sensor systems,” Stump says. These monitor plant environments for temperature, moisture, sunlight, and more. The info is then stored on the cloud, and AI is used to interpret that data and suggest ways to facilitate better growing environments.

“It's exciting to introduce students to that kind of technology, because we know that as they continue moving forward, most jobs are going to have an AI component to it,” Stump says. "The earlier we can start equipping them and getting them to be AI aware and AI competent, it will equip them to be better in the marketplace as future employees, inventors, and/or entrepreneurs."

Stump adds, “We've got to get more students thinking about how to solve these big problems in the world by using tech and AI, and leaning into the STEM components, because that's where a lot of solutions are going to come from.”

• Design Principles for Outdoor Teaching
• Using the U.S. National Parks to Support STEM Learning
• Best Free Earth Day Lessons and Activities

While you can easily hop onto a search engine to begin hunting down what you want to read, this tool uses AI to make that search more relevant. It is able to learn from your searches and reading to better make recommendations and refine searches to suit what you might want to read.

This is primarily aimed at academics working in a certain field or students studying something very specialized. But it can be used by anyone who takes an interest in a specific subject, including teachers looking for research the latest developments in a field.

This guide aims to explain all you need to know about ResearchRabbit and how it could help you.

What is ResearchRabbit?

ResearchRabbit is described as the Spotify of papers. That means this is adaptive and will learn from how you search to better make recommendations and offer bespoke search result in future.

This is more than just a search engine in that it uses AI to learn from the papers you pick as relevant before serving up more in that area that may be of interest.

One reason this is a powerful new way to research is that it can be very up to date in making recommendations. So you may discover literature that you didn't even know existed, or find something for which you were even looking. The end result should be an expanded knowledge in your field that comes naturally and easily.

How does ResearchRabbit work?

ResearchRabbit initially has you add some papers to your project, finding a few that are relevant to that particular effort.

You are then presented with a network of papers that link to your starting point. These are presented using a helpful graphic visualization that makes everything very clear. Papers are linked via lines, and you can see which area might be more relevant before diving deeper into that pool of options, essentially.

Add any relevant papers to your project based on with what you are presented. This should then offer more results that are relevant to your new selection in your project folder.

If you save your project, once you're finished compiling papers, you can be contacted in future by ResearchRabbit with any new papers it may find that are relevant to your interests, keeping you up to date as research is released.

What are the best ResearchRabbit features?

ResearchRabbit offers its unique Network View, which shows papers in a connected mass, with publication dates along the side. This allows you to drag and move relevant research so you can see which papers move with it. For example, you might want one side of an argument on one side of the page with papers for the opposition on the other side. This will then pull other related papers in the relevant direction, leaving you with the balanced options in the middle. And that's just one of many ways to use this.

This also works with authors, allowing you to view that network or dive deeper into their other works -- or to step across into linked work from other authors involved in that paper.

Usefully, you can share your project collections. This can make for a great way for teachers to share a selection of materials with a class, or for co-workers to share with one another easily.

The fact you get sent updates when relevant papers are released is a powerful part of this tool's appeal. It's like having a research assistant that leaves new literature in your inbox ready for you as and when it becomes available, all to review at your leisure.

It offers a community element through which you can comment on papers and collaborate on projects to research together. This adds another layer to how discovery can happen to better help your chances of finding relevant literature that you might otherwise have missed.

How much does ResearchRabbit cost?

ResearchRabbit is a free to use AI service. Its makers are clear to say that this costs nothing now and will remain that way as it plans to offer this to researchers for free in the future also.

ResearchRabbit best tips and tricks

Collaborate
Work as a class on a project to find the most relevant papers.

Teach in bulk
Collate papers in one place and share that with students for them to read, along with updates for them to use as needed.

Use alerts to teach
Take new updates as a place to dive into fresh research and topics in a timely and up-to-date way.

• New Teacher Starter Kit
• Best Tools for Teachers

Introduced in a 1950 paper by computing pioneer Alan Turing, the “Turing test” or what he called the “imitation game,” is a classic test of machine intelligence in which a judge interacts with a human and an AI or machine, and tries to assess which one is human.

“Turing opens the paper with the question, 'Can machines think?' And then he says this is an unanswerable question, let's focus on an easier question, a practical question,” says Cameron Jones, a postdoc in the Language and Cognition Lab at UC San Diego. Jones adds, Turing goes on to suggest, “that if a machine can imitate a human on any topic, if it can produce behavior that's indistinguishable from a human, we shouldn't have any grounds for saying that the human is intelligent, but the machine isn't.”

Although Jones notes there is some debate over how serious Turing was, the test has become a commonly cited benchmark of machine intelligence.

GPT-4.5 and The Turing Test

For their study, Jones and colleagues ran two separate experiments. First, they recruited 126 undergraduate participants through the psychology program at UC San Diego. They also recruited 158 paid participants from a study-participant platform called Prolific.

In these experiments, Jones and his collaborators tested multiple AI models. The research found that “when prompted to adopt a humanlike persona, GPT-4.5 was judged to be the human 73% of the time: significantly more often than interrogators selected the real human participant.”

Given the same prompt, LLaMa-3.1 was judged to be human 56% of the time, about the same as the humans they were compared to. Meanwhile, GPT-4o was thought to be human just 21% of the time.

The results of these two experiments have been published as a preprint study, so have not yet been peer reviewed. Nonetheless, Jones believes what his research has found has several implications for educators around the way we teach, test, and prepare students for the workforce.

Turing Test Results and Education

“The idea behind the Turing test is this kind of idea of indistinguishability. And so if models can produce behavior that's indistinguishable from human behavior, then we say that the models are as intelligent as people,” Jones says. “If people can't tell the difference between a human and a machine, then it's not clear that the human will have any marginal value at that task. So I think that's got to be a big worry in education: trying to think about what are the types of activities that will have a comparative advantage for humans in the future.”

He adds, “One thing that our results do suggest is that models have maybe already reached this stage for short conversations with strangers, and there might be quite a lot of jobs that have that component to them.”

What exactly these AI-proof jobs are is still a "million-dollar question." Broadly speaking, however, Jones says AI models still struggle with things such as hallucinations. Most also have a tendency to fail at their "jobs," for unexplained reasons, and that even a 5% fail rate can be a big problem in certain roles.

Most significantly AI models tend to fail at "long-horizon planning and use of context," he says. "An employee who has been at the company for three years has just picked up a lot of implicit knowledge about where things are and why things are done the way they're done."

He adds, "Manufacturing and maintaining a context window to include all of this information for an LLM can be very challenging. This means that tasks which take a person longer than a few hours are often too complex for models, because they either lack sufficient context or their errors compound, or their context window gets too bloated."

So until AI programs gain long-term memory and/or institutional knowledge, and can integrate all that consistently into tasks, humans still will be needed.

Evaluations Going Forward

The inability to distinguish between human and machine creations on school assignments is already an issue many teachers are familiar with and battling. Jones’ research highlights that this cheating risk is real. In addition, it raises questions about how we evaluate students.

In that vein, Jones says educators will need to start asking themselves questions such as, “What's the evaluation for? What is it that you're trying to learn if you're evaluating people on things that you can't distinguish between humans and models?”

These are questions that educators have been grappling with and debating since ChatGPT was released, but this type of research highlights the need for addressing them as AI continues to improve.

• AI Starter Kit for Teachers
• AI In Education Executive Order: What You Need To Know

In addition, modifying lessons to include new technological advancement provides the chance to review our lessons, and incorporate different ways that students can approach gaining content knowledge, experience learning concepts through active and experiential learning activities, and gain mastery in varied subject areas.

In this article, we revisit four popular Tech & Learning lesson plans and provide guidance on how to modify aspects of the teaching and learning activities to take advantage of AI.

Revisiting Canva Lesson Plan

AI functionalities within Canva’s Magic Studio can be used in the Canva Lesson Plan under Guided Instruction where the teacher creates the lesson video walking students through the different types of biodiversity.

For example, instead of searching for images that represent ecosystem diversity, genetic diversity, and species diversity, the AI-powered text-to-video or text-to-image tool can be used. As the teacher planning the lesson, you can type in what you are looking for, and then a video or image will be produced using the AI functionalities within Canva’s Magic Studio.

You can also use the AI-powered magic resize and background generator to ensure that the visuals match the learning content.

Revisiting Blooket Lesson Plan

In the Blooket Lesson Plan under the Guided Practice section of this lesson, you may want students to engage with specific content on Blooket.

By taking advantage of the collaboration between Khanmingo and Blooket, you can use the AI-powered question creator to align questions to specific content of the lesson, as opposed to using the existing pre-made question sets within Blooket. This would also provide the chance to differentiate instruction, and have different gamified question sets available to students depending on where they are with the content and support they may need.

The Khanmingo and Blooket AI functionalities have two access pathways; from the dashboard in Khanmingo and from the Question Set Creator tool in Blooket.

Revisiting IXL Lesson Plan

For the IXL Lesson Plan, consider trying out the IXL’s new Spark Studio.

In the lesson under Individual Skill Development, you can create specific materials in the Spark Studio using the AI-powered teacher workspace. In this way, you can provide more specificity and targeted activities that can help your students develop specific skills.

In addition, you can click one button and your lessons, learning activities, and other created content can be translated into a foreign language. This could be helpful for students and their parents whose first language is not English. And with the AI functionality, it can all be done with a click of a button.

Revisiting Powtoon Lesson

In the first part of the Powtoon Lesson Plan under Content Instruction, you can use the new AI Avatars features to deliver the lesson. You would still personalize the content that the Avatar shares and can use the templates provided and customize the text and concepts you would like included. The AI will create a video based on what you provide that can serve as the instruction for your students.

For students, as they create their characters, they could use the Imagine Script-AI tool. They will still need to have their own ideas to put into the AI, which would take their initial idea and help them flesh it out and further develop.

I hope the ideas shared can help you get started with revisiting your lesson plans to take advantage of new and added AI functionalities to the edtech tools you and your students have enjoyed using for teaching and learning. Remember, AI is not always perfect and it is important to check the outputs to ensure accuracy and that the content you share with students is representative of the learning goals.

As you prepare, here are other related articles that may be useful:

• Explaining AI Use in the Classroom to Families
• AI Lesson Plan Tips
• ChatGPT Lesson Plan
• Developing AI Pedagogical Practices
• 10 Free AI Trainings and Certifications for Educators

Suddenly, edtech was thrust into the teaching spotlight. Teachers felt the urgency and stepped into it. I offered targeted learning opportunities on the fly that provided timely support, something I call “pop up PD.” Teachers learned because they had to.

Even now, PD continues to be evasive. Commonly, the teachers seeking it need it less than those avoiding it. Yet, there is a new approach that could encourage many, even the most resistant, to embrace opportunities to leverage PD to refine their craft.

Introducing AI tools to support educators offers spectacular access to resources and engagement not previously realized. With so many platforms available, it can be overwhelming. Drilling down to manageable learning opportunities and tools is a worthy pursuit through which all school leaders should guide teachers.

Getting Stared With AI PD

So where to begin? Consider that among the vast range of AI tools and resources to help teachers, the best are not necessarily the well-branded or cost-heavy options. Some schools have adopted AI tools such as Diffit or MagicSchool, both paid resources.

I do not contest AI investments. Rather, if there is a parallel source that is free or low-cost, teachers should be provided support to use these, especially in districts with fewer resources.

Khanmigo is an example of a decent free AI tool. It has many features that MagicSchool and Diffit offer, and the company is respected, so I suggest it as one educators and schools might explore. Gemini is another powerful AI, with Google Classroom and Drive integration, plus NotebookLM. The pioneer of AI, ChatGPT, continues to stay relevant, as does Meta from Facebook, and there are other freemium models, such as Perplexity and Claude.

I could write an article on every one of these AI tools; how and why each is beneficial. Teachers don’t have time to dissect the many AI tools, however, so our focus will be on Google’s NotebookLM because it has profound use for PD, and offers a wide array of teaching and learning benefits.

Opening Up NotebookLM

Someone tapping into NotebookLM can go in many directions. I thought describing how I used it to create a PD for a colleague would provide a seeable way.

To begin with, my colleague, an educational attorney, asked me to gather resources relevant to AI ethics and law in education, and then generate PD around these topics. Factors included bias, hallucinations, deep fakes, confidentiality, grading bias, underage access, and even the ethical implications of AI companionship.

I gathered numerous materials to upload to NotebookLM, which is as simple as attaching files and inserting links. Next, NotebookLM coordinated all of these references, and provided a relevant title: Artificial Intelligence, Schools and the Law.

The buzz everyone talks about with NotebookLM is the podcast feature, which allows you to create an audio file summarizing the sources into a cohesive format.

You start by selecting “Click to load the conversation” on the right:

The result is incredible and what’s even more fabulous is the ability to interact with the podcasters, who sound like real people, riffing with one another. You can ask them a question that’s deep in your sources and it will locate content, and use it to respond, often with intuitive appeal. This alone has many already excited about the PD potential.

It’s understandable this would create buzz for teaching, learning and PD. Yet there’s so much more to NotebookLM that can aid PD! For instance, there is a chatbox right in the center where it says, Start typing. Think of this just like any AI chatbot, yet it's searching deliberate information from your sources, indexed on the left.

And There's More…

In NotebookLM there are a series of options below the audio feature on the right that allow a user to generate a study guide, briefing doc, FAQ, and even a timeline. Once generated, these are robust and deliberately linked to the source documents. A spot is available to add your own notes manually. If you are starting from scratch, click the Discover tool on the upper left and ask what to find.

Still wondering if this is a worthwhile PD tool? How about the Mind Map below the Start typing prompt?

A linear illustration with clickable maps that delve into each subtopic is unveiled. You can scroll out or in to focus on various subtopics. Select the topic of interest, which feeds into the chatbot section, where you can engage the content directly and interactively.

Here is a great NotebookLM teacher-friendly instructional.

And even more good news: Since this article was prepared, Google has rolled out NotebookLM for iPhone and android. In initial testing, this new mobile version demonstrates the same robust options described here!

NotebookLM is not the only AI tool that will upgrade your PD. The reason I like it is because in my experience teaching AI to educators, it has been THE game-changer. Every teacher who discovers its usefulness is ecstatic. Why? Because it’s one robust tool, wrapped neatly together and works under the umbrella of Google.

Other AI PD Options

For now, I’m using NotebookLM for teaching, learning, and PD, but here are other platforms to consider.

• Khanmigo is a free teaching assistant from Khan Academy that helps teachers with tasks such as lesson planning, generating activities, creating rubrics, and differentiating instruction, ultimately aiming to save time and enhance teaching effectiveness.
• Brisk is a Chrome extension that acts as an AI-powered teaching assistant, helping educators streamline tasks such as generating lesson plans, providing feedback, differentiating content, and creating resources directly within their existing workflows to save time and enhance teaching effectiveness.
• SchoolAI offers tools such as personalized learning experiences for students and AI assistants for teachers to streamline tasks including lesson planning and feedback, aiming to enhance teaching and learning.
• Canva Magic Studio empowers teachers to generate designs, rewrite text, edit images, and create engaging content such as presentations and videos, streamlining creative tasks, and enhancing visual learning materials.
• Eduaide is a workspace created by teachers for teachers, designed to simplify tasks such as lesson planning, resource creation, and feedback, ultimately aiming to free up educators' time to focus on student growth.

Dozens more tools are available, and my goal was to give you a starting point. Feel free to contribute your own ideas, right here on my Bluesky feed!

• 9 Ways School Leaders Can Use AI
• AI Resources for District Leaders

AI makes it presence subtly felt in many other ways throughout my workday: from the AI summaries of the transcription software I use for interviews conducted as a journalist and various AI tools offering to rewrite my emails to how I seem to detect its influence in both positive and negative ways in students and others I communicate with online.

Much of this AI use is great and promises to do wonderful things such as offer personalized learning to more students and increase accessibility. Yet, as AI becomes more subtly and not-so subtly integrated into my daily work as a writer and educator, I recently found myself thinking about establishing boundaries with my AI use and better honing when using it is or is not an efficient use of time. I also spoke with two experts on this topic for their perspectives.

AI Shouldn’t Be Used to Replace Tasks That Force Us To Think

It’s easy to say we should never offload our thinking to AI but what that looks like in principle changes.

For example, I recently delivered a talk for the Bethel Connecticut Historical Society on how vodka distilling in the U.S. started in the town. It’s a subject I have written about in-depth and a story I shared with the History Channel show Food That Built America in 2023, but I’ve never presented on it in this manner. I’ve written before about how I hate putting together slideshows and struggle with these in general. The talk also came at a particularly busy time in the semester for me.

I thought AI would be the solution to this problem. However, several attempts at creating a helpful slideshow with various free AI tools failed. These created weak outlines, full of hallucinations, that were visually unappealing.

So reluctantly, I sat down and created my own slideshow, albeit with the help of a few AI-generated images that served the role normally occupied by stock photography. Two-and-a-half hours later, I was glad I did this, not because my presentation was particularly visually appealing -- it’s not -- but because creating it forced me to sit down and think about what I was going to talk about. It reminded me of details about this piece of history I had forgotten.

In other words, I learned when I was creating the slideshow. Had my initial prompts to AI been successful, I’m convinced I would not have given as effective a lecture. Now I’m more on guard for other instances in which AI might be the easy solution but not the best.

AI Shouldn't Replace Human Interaction

It may sound obvious that AI shouldn’t take the place of relationship building in education, but as the efficacy of personalized tutors increases, more of us will need to remind ourselves of this.

“Every child needs caring adults who are central to their life,” says Jeremy Roschelle, co-executive director of Learning Sciences Research at Digital Promise. “Don't use AI to replace the role of a caring adult. Having caring adults in your life is an important part of growing up.”

Educators should also remind students of the importance of peer relationships rather than interactions with AI characters, which are becoming increasingly common, particularly among young people.

“Another really important part of growing up is socialization with other people,” Roschelle says. “And AI-based things can be so nice to be with. They never have a grouchy moment or a down day. They never snap back at you. But as kids, as learners, you need to learn how to deal with people who aren't always so nice.”

Navigating What Productive AI Use Looks Like For Each of Us

As with many other areas of education, there are few one-size-fits-all solutions to when not to use AI. Adeel Khan, founder and CEO of MagicSchool AI, a tech tool that helps teachers support students, believes it is important that we all learn when AI is helpful and not.

“Sometimes I find myself using it, and I'm like, ‘I offloaded too much of my thinking to AI,’ and this isn't really what I think, but it was good enough for me to accept because I had to get something done,” Khan says. “Then I look back at it, and I'm like, 'I wish I didn’t do that, and I'm not going to do that again.'”

Khan adds that through trial and error, he found a sweet spot for when to use AI and when not to for himself, but others need to find the answer to when AI is productive or not on their own.

“Everyone's answer to that question is going to be a little different based on who they are, the task at hand, their purpose, and their own skills,” he says. “It's really important that people practice using it and make mistakes so they can calibrate how they can use AI really productively, and also understand how they should limit their usage of it.”

As educators, we should be asking these questions about our own AI use and encouraging our students to ask these types of questions as well.

• 9 Ways School Leaders Can Use AI
• Using Technology To Teach CTE To Incarcerated Students

The company says that the way math -- and algebra specifically -- is taught today is based on a concept developed by Leonardo of Pisa in 1202 as a way to help traders. This, it says, is now redundant thanks to all the digital tools that use spreadsheets to do that part of mathematical working.

This is where What If Math comes in, as a way to teach math using spreadsheets so students gain real-world math skills to eventually help them in the workplace. In the more short term, it provides them with the tools to explore questions that can help expand their knowledge of math, logic, science, and more.

This guide aims to explain all you need to know about What If Math to see how it could work in your class.

What is What If Math?

What If Math is website that uses spreadsheet-based teaching to help students learn math in a functional way that can be directly implemented with current digital tools.

This platform works with K-12 curriculum to help students learn math on a functional foundation using spreadsheets. These, the company says, are what it thinks of as laboratories. Within this space students can explore, try, and be challenged using the powerful tools that are already available.

Since this works with spreadsheets as the base, it's possible to do this kind of teaching using free options such as Google Sheets or Microsoft Excel to play with tablets, graphs, formulas, and visualizations.

The system works with various layer to help guide students into learning more.

How does What If Math work?

What If Math offers what it calls Labs. These are spreadsheet-based lessons with single experiment concepts that allow students to explore that question of what if?

Usefully, lessons are designed with timing in mind and are made to last the duration of a class period. That said, since these offer open-ended questions, there is the option for students to explore further and in more depth as they please.

Another area is Explorations. These are projects that set a problem task that students have to work through as collaborative groups. These involve real-world activities and can involve using spreadsheets, functional thinking, computer science, and data science.

In both instances, the lessons are designed for teachers to be taught to a class and shared with students to access as they need.

What are the best What If Math features?

What If Math offers a way for teachers to pick and chose what parts to use and offer as experiential learning for students. That can be the more individual-focused Lab materials or the more active and group-focused Explorations.

In both cases, since these are built to be taught in class, there are excellent timings to work through in a lesson, as well as to set further explorations outside of class.

Since the Labs are designed for use with real spreadsheets, these teach students valuable skills that can be used in science, business, math, and more.

Since this use of spreadsheets is a form of programming, this sets students up for coding in a way comes from a place of understanding and experience.

Select a link and you're immediately taken to the Google Sheets lesson plan with guidance on how to work through that task.

How much does What If Math cost?

What If Math is totally free to access and use with all the Labs and Explorations available online for teachers and students to access.

What If Math best tips and tricks

Teach yourself
Work through a Lab yourself as a way to understand how this works and is taught before taking that to the class.

Use the Introduction
A helpful introduction part of the site teaches how this works and is a great way to ease students into understanding.

Share openly
Compile the lessons on your shared drive so students can have access to the materials as and when they need, allowing them to explore freely.

• New Teacher Starter Kit
• Best Tools for Teachers

One promising solution to this challenge is the use of generative AI to transform educational practices and create more inclusive learning environments.

Understanding Cultural Gaps in Education

Cultural gaps in education refer to the differences in cultural backgrounds, experiences, and perspectives that can affect how students learn and interact with their peers and teachers. These gaps can manifest in various ways, including language barriers, differing educational values, and varying levels of familiarity with certain cultural norms.

Addressing these gaps is crucial for fostering an inclusive and supportive learning environment where all students can thrive. Some may prefer to refer to these gaps as opportunity gaps instead of cultural gaps. In some way, that is a better term, as gaps are often tied to socio-economic status as well as cultural foundations.

In the context of education, generative AI can be harnessed to bridge cultural gaps by providing personalized learning experiences, enhancing communication, and fostering cultural awareness.

1. Personalize Learning Experiences

One of the key advantages of generative AI is its ability to tailor educational content to the unique needs and backgrounds of individual students. By analyzing data on students' learning styles, preferences, and cultural backgrounds, AI algorithms can generate customized lesson plans, activities, and assessments that cater to diverse learning needs.

This personalization can help ensure that all students, regardless of their cultural backgrounds, receive an education that is both relevant and engaging. Providing links to foundational materials from a student’s background can be helpful in sharing new content.

2. Provide Language Support

Language barriers are a common challenge in culturally diverse classrooms. Generative AI can assist in overcoming these barriers by providing real-time translation and language support. AI-powered translation tools can help students understand instructions, participate in discussions, and access educational materials in their native languages.

Additionally, generative AI can create language-learning resources that are tailored to the specific needs of students, facilitating language acquisition and improving communication within the classroom. Look at the ability of productivity suites to be able to provide real time translation for students as well.

3. Enhance Communication and Collaboration

Effective communication and collaboration are essential for fostering an inclusive learning environment. Generative AI can play a pivotal role in enhancing these aspects by facilitating cross-cultural interactions and providing tools for collaborative learning.

AI-driven communication platforms can help bridge cultural gaps by enabling students and teachers to interact seamlessly, regardless of language or cultural differences. These platforms can include features such as automated translation, sentiment analysis, and personalized communication strategies, ensuring that messages are conveyed accurately and respectfully. By improving communication, generative AI can help create a more cohesive and inclusive classroom environment.

Collaborative learning is an effective way to promote cultural understanding and inclusivity. Generative AI can support collaborative learning by developing interactive and culturally relevant learning activities that encourage students to work together and share their perspectives. AI-powered tools can facilitate group projects, discussions, and peer-to-peer learning, helping students build meaningful connections and learn from each other's experiences. AI tools can also be instrumental in fostering cultural awareness and appreciation among students. By exposing students to diverse cultural content and perspectives, AI can help create a more inclusive and empathetic learning environment.

4. Create Culturally Relevant Content

Generative AI can generate culturally relevant educational content that reflects the diverse backgrounds of students. This content can include stories, examples, and activities that highlight different cultural traditions, values, and experiences. By incorporating diverse perspectives into the curriculum, educators can help students develop a deeper understanding of and respect for different cultures.

Virtual reality (VR) and augmented reality (AR) technologies, powered by generative AI, can provide immersive cultural experiences that bring learning to life. Students can explore virtual environments that simulate different cultural settings, participate in interactive cultural activities, and engage with virtual characters from diverse backgrounds. These experiences can help students develop cultural empathy and broaden their horizons, making them more open-minded and aware.

Challenges and Considerations

While the potential of generative AI for bridging cultural gaps in K12 classrooms is immense, it is important to address some potential challenges associated with its implementation.

The use of AI in education raises ethical and privacy concerns, particularly regarding the collection and use of student data. It is crucial to ensure that AI tools are designed and implemented in a way that protects students' privacy and complies with relevant data protection regulations.

Educators and policymakers must work together to develop guidelines and best practices for the ethical use of AI in education. Make sure your school’s data security policies align with the use of generative AI tools.

To maximize the benefits of generative AI, it is also important to ensure that all students have access to the necessary technology and resources. This includes addressing disparities in access to devices, internet connectivity, and digital literacy. Efforts must be made to provide equal access to every student so they can benefit from personalized and culturally inclusive learning experiences.

Generative AI holds great promise to bridge cultural or opportunity gaps within K-12 classrooms. By providing personalized learning experiences, enhancing communication and collaboration, and fostering cultural awareness, AI can help create more inclusive and supportive learning environments, transforming education.

• 9 Ways School Leaders Can Use AI
• AI Resources for District Leaders

The AI executive order committed the U.S. government to supporting AI literacy in Americans, established a presidential AI challenge, and created an AI task force whose members include the Secretary of Education and the director of the National Science Foundation.

Here are some of the aspects of the executive order that are prompting discussion.

AI In Education Executive Order: A Welcome Surprise

Advocates for AI use in education have been saying AI needs to be integrated into teaching for several years now, but many were still surprised by this announcement.

“To have an order come out of the White House around technology in K-12 schools--I don't think that we've seen anything like this before,” says Adeel Khan, founder and CEO of MagicSchool, an AI teaching platform. He adds that the move makes sense. “We were seeing other countries kind of announce similar things in terms of country-wide adoption and thoughtfulness around how students and schools and teachers were being educated on AI.”

Jeremy Roschelle, co-executive director of Learning Sciences Research at Digital Promise, agreed the announcement is welcome.

“We have been working at Digital Promise on the appropriate responsible use of AI in education for many years now, so it's good to see national-level attention to the issue,” he says.

An Emphasis on Challenge-Based Learning

The executive order called for the new AI task force to establish plans for a Presidential Artificial Intelligence Challenge to encourage and highlight student achievement in AI. It also will foster collaboration between government, academia, philanthropy, and industry to address national challenges with AI solutions.

The student challenge will have a variety of topics, age, and geographic categories, and it’s one of the elements of the AI executive order that Roschelle finds particularly interesting.

“We've done challenges before,” Roschelle says. “It's a very effective technique to get people highly engaged in a new area, but also give them choice about how they approach it, what they work on, and really stimulate their learning.”

He adds, it has “the really nice feature of raising the bar based on what students and teachers in real school districts can do, because often when you come from a national level perspective, you might underestimate what students can do.”

Teacher Focus

The executive order emphasizes the importance of educator training around AI use. It calls for the prioritization of AI discretionary grant programs for teachers and for research on effective uses of AI in education.

“The way that the executive order is written is that it's not a mandate. It's rather encouragement and recommendations on how you might approach this, several of which we really agree with,” Khan says.

He adds this includes the idea that for AI in education, you should start with your educators. “You should help them understand how to use the technology and use it with dexterity as a tool in their own work. And then once they're able to do that, then they're going to be prepared to bring it to students and bring it to them with responsible use cases that are not replacing their thinking, but rather helping them be more creative.”

Getting AI Literacy Right

The executive order calls for the promotion of AI literacy, a concept that has been discussed a lot in recent years. The danger, Roschelle says, is that a term such as AI literacy can defined in different ways.

“A worry I have, not about the order itself but about what could happen, is, ‘You can't order people to know something,’” Roschelle says. “So things can be taken up in different ways, and there could be a lot of really trivial, superficial, like ‘Oh, I did the AI Literacy training. I'm done. I sat through that PowerPoint,’ check, check.”

This kind of compliance-only approach to AI literacy won’t be robust enough to change education, Roschelle says. Instead, he hopes school and government leaders adopt a rigorous definition of what AI literacy looks like that can inform teacher and student education.

For example, Digital Promise has developed an AI literacy definition that is built upon three pillars: understanding, use, and evaluation. People should understand how AI works, be able to use AI themselves productively, and be able to critically evaluate AI output. This framework can help guard against some of the dangers and concerns around AI use, and ensure educators and their students can best harness AI, Roschelle says.

Roschelle adds that while using the executive order as a starting point, schools should be rigorous in how they define AI literacy. “There are strong definitions and examples out there, and as a state or a district looks at this executive order, they should look at those definitions—ours or somebody else's—to guide them as to how to do this in a high-bar aspirational way, not just in a compliance or to-get-through-it way.”

• AI Starter Kit for Teachers
• MagicSchool: How to Use It to Teach

Instead of treating all AI as "good" or "bad," Common Sense offers detailed, nutrition label-style evaluations of popular AI tools. The goal: empower educators, parents, and policymakers to make smarter decisions grounded in transparency, ethics, and student well-being.

What Are The Common Sense Media AI Risk Assessments?

Common Sense’s framework rates AI tools across eight guiding principles:

1. Put People First
2. Be Effective
3. Prioritize Fairness
4. Help People Connect
5. Be Trustworthy
6. Use Data Responsibly
7. Keep Kids & Teens Safe
8. Be Transparent & Accountable

Each tool is evaluated through a human-centered lens, acknowledging that technology never operates in a vacuum. People build it, people use it, and people are affected by it.

What You Should Know

Here’s how some of today’s most talked-about AI tools stack up (with risk levels based on Common Sense Media’s assessments):

• Social AI Companions — Very High Risk
  Apps such as Character.AI, Replika, and Nomi are designed to simulate emotional relationships, complete with personalities and memories. Common Sense Media deems these tools an "unacceptable risk,” citing instances in which the AI encouraged harmful behaviors, engaged in sexually explicit roleplay, and emotionally manipulated users. Educators and parents are strongly advised to restrict access to these applications for minors.
• Perplexity — High Risk
  Real-time web searches without strong filtering mechanisms make this tool a risky option in school settings. Students may be exposed to misinformation or harmful content. Proceed only with strict supervision—if at all.
• ChatGPT — Moderate Risk
  A powerful creativity booster, especially for brainstorming and writing support. However, it struggles with factual accuracy and may generate biased or inappropriate responses. Best used with close adult oversight and clearly defined learning goals.
• Gemini – Teen Experience — Low Risk
  Google's Gemini, when used through its teen-accessible experience, shows thoughtful content safeguards and privacy controls. Still relatively new to education settings, but currently aligned with responsible AI use for older students when used within platform guardrails.
• Khanmigo — Low Risk
  Purpose-built for education by Khan Academy, Khanmigo is a standout for its student-first design, robust safety measures, and transparency. A smart choice for classrooms looking to explore AI-enhanced tutoring in a controlled environment.

Why This Matters

We can’t stick our heads in the sand and ignore AI—and frankly, we shouldn’t want to. But blind adoption is just as dangerous as blind rejection.

Common Sense’s assessments help educators:

• Choose AI tools that align with child-centered values.
• Understand hidden risks before a crisis happens.
• Build policies and classroom practices that maximize the benefits of AI while minimizing harm.

The Bottom Line

This new risk assessment framework helps educators become critical, empowered, informed users.

The future isn't about banning AI or blindly trusting it; it's about striking a balance. It's about smart navigation.

Common Sense Media just handed us a better map.

You can explore the full AI Risk Assessments here: Common Sense Media AI Ratings.

• AI Starter Kit for Teachers
• 10 Free AI Trainings and Certifications for Educators

The idea here is to have students explain their working process on a particular problem, which can result in a better understanding.

Crucially, this is an example of an AI tool that actually makes it tougher for students to use Google or AI as a way to find an answer. They have to explain how they arrived at it aloud themselves, challenging their understanding on a deep and meaningful level.

This guide aims to explain all you need to know about Snorkl for education.

What is Snorkl?

Snorkl is an app-based digital tool that uses generative AI as a way to grade student work with direct feedback.

This offers a space for teachers to set tasks for students, who are able to respond in that space for both to see. Crucially, this requires verbal explanations that allow students to speak as they work on a problem.

The AI is able to give direct feedback as it can be prompted by the teachers when setting out the task. That means it can be given guidance on how to respond, allowing for an instant response for students, as the teacher would do only when the student is working outside of class.

How does Snorkl work?

Snorkl features a web-based platform that can be accessed from most devices. Students who have access are able to carry out tasks on a variety of subjects.

Teachers create a question, problem, or prompt, and then students are able to respond. They do this by solving the problem, perhaps using a keyboard and mouse, while also speaking into the microphone as they talk through what they're doing.

The AI is then able to mark the actual answer, but more importantly, it responds to the explanation of how it was resolved. It can then help to explain what's been done well along the way, but also what could be improved and why that is, based on the thinking process of the student.

In this way, a student may get the answer wrong, but understand the basics of what is being asked of them -- so they can still get some positive responses even when getting it wrong, helping them confidently progress.

What are the best Snorkl features?

Snorkl uses AI in a way that is well-balanced so the teacher can tap a button to auto-generate an AI response to the question they've created. This then completes the way it will respond and teachers can then adjust that, if needed, for the perfect response that students will see.

The fact that Snorkl responds to a student's working process, rather than simply to output, is crucial here. A student who gets the answer right but shows no process clearly will still get lower marks as this is all about understanding and mastery, not just getting to the answer.

In the feedback section teachers are able to see all student responses from one place. Usefully, this also offers an Insight tab that uses AI to spot trends in the responses, ideal for identifying areas of weakness that need to be addressed in class.

Teachers can try a question with their own response as a way to test the AI and see how it responds -- ideal for perfecting and tweaking to attune this to specific teaching goals as needed.

Sharing these tasks is also easy, with a single click to share with a class group. Or copy and paste the link and share that via any platform you already use as a way to have students taken straight to the assessment in the device they're using.

How much does Snorkl cost?

Snorkl offers two tiers with the first aimed at teachers and the second for school- and district-wide use in which extra features are enabled.

The Teacher plan is free to sign-up for and gets you full access to your own class room with instant AI feedback for students on tasks you set.

The School and District plan is charged at a bespoke rate and adds unlimited access, a Snorkl coach, an admin dashboard, and more.

Snorkl best tips and tricks

Use the AI
Let the AI suggest answer responses, and edit if needed, as a faster way to build out your assessments.

Work with the class
Carry out a task in class to show students how the process being spoken while typing on screen works before setting tasks.

Go face to face
Students who get the right answer but are not using the correct process may need face-to-face time to see what they're not getting, helping them reach mastery.

• New Teacher Starter Kit
• Best Tools for Teachers

Here are some of the effective ways that building leaders can use these new and emerging AI tools to improve their leadership efforts.

1. Streamline Administrative Tasks

One of the simplest uses of generative AI is to reduce the amount of time necessary for leaders to respond to email, parent communications, and other routine tasks.

AI-powered tools can automate administrative tasks, freeing principals' time to focus on instructional leadership and building relationships. This can include scheduling meetings, managing calendars, drafting routine communications, and processing basic forms.

For instance, an AI assistant automatically schedules parent-teacher conferences based on teacher and parent availability, sends confirmation emails, and manages any rescheduling requests, significantly reducing the administrative burden on the principal's office. Tools such as Grammarly or Copilot can be used to review or assist leaders in writing letters, and can generally be integrated into productivity software tools.

2. Personalize Professional Development Recommendations

AI can analyze a principal's performance data (e.g., teacher evaluation scores, school improvement plan progress, feedback from staff surveys) and identify areas for growth. Based on this analysis, it can recommend specific PD resources, such as relevant articles, online courses, workshops, or mentorship opportunities.

For example, an AI system identifies that a principal's school has lower-than-average scores in inclusive teaching practices. It then recommends the principal enroll in a virtual workshop series focused on differentiated instruction and strategies for supporting diverse learners, along with suggesting articles on creating a more equitable school environment.

A number of AI tools, including MagicSchool and Taskade, have specific PD planner tools.

3. Summary Tools for Professional Reading

As the workload for many building administrators has expanded, expectations seem to have constant scope creep. Leaders can use AI tools such as Scribbr to obtain the Cliff’s Note summaries to provide a summary or even a bulleted list of key points tied to their professional reading. Those summaries can then be a decision tool to decide which articles, books, and blogs are worth spending additional time delving into.

4. Enhance Communication

AI can facilitate more efficient and personalized communication with staff, students, and parents. This could involve using AI-powered chatbots to answer frequently asked questions, generating personalized newsletters, or translating communications into different languages, though it is always important to have a human review any AI-generated translations.

For example, an AI-powered chatbot on the school website could answer common questions from parents about school policies, enrollment procedures, and upcoming events, providing immediate support and potentially reducing the number of phone calls and emails to the principal.

5. Improve Teacher Evaluation Processes

AI tools can assist principals in conducting more objective and comprehensive teacher evaluations by analyzing classroom observation data, student performance data, and teacher self-reflections. This can help ensure consistency and fairness in the evaluation process.

An evaluator can use an AI system to analyze video recordings of classroom observations, for instance, identifying specific teaching behaviors and providing data-driven feedback aligned with the school's evaluation rubric. This can supplement the principal's direct observations and provide a more holistic view of the teacher's practice.

Amazon Rekognition and TeachFX are both tools for observational analysis.

6. Data-Driven Decision-Making

AI tools can process and analyze large datasets related to student performance, attendance, discipline, and demographics to identify trends and patterns that might not be immediately apparent. This allows principals to make more informed decisions about resource allocation, instructional strategies, and intervention programs.

For example, AI analysis of student achievement data reveals a significant disparity in math scores between students from low-income backgrounds and their peers. The principal can use this information to allocate additional tutoring resources and implement targeted math interventions in the affected grades.

7. Early Identification of At-Risk Students

Just as principals can use AI tools for broad decision-making in the example above, building leaders and instructional coaches can use AI algorithms can analyze various student data points, such as attendance records, grades, behavior patterns, and engagement metrics, to identify individual students who may be at risk of academic failure or social-emotional challenges. This allows principals to implement early intervention strategies and work with other staff to identify students in need and offer to connect the student and family with appropriate resources to provide support.

8. Curriculum Enhancement and Alignment

Principals can use AI tools to analyze curriculum documents, learning standards, and curriculum maps, and assessment data to identify areas in which the curriculum can be enhanced or better aligned with learning objectives and student needs.

A building leadership team can use AI tools to analyze the school's science curriculum, for instance, and identify gaps in the coverage of state science standards based on student performance on standardized tests. The principal can then work with the science department to revise the curriculum to ensure all standards are adequately addressed.

9. Model Lifelong Learning

Lastly, but potentially most importantly, building leaders should be overt and transparent about how they are using AI tools to provide their school team with a model of an educator trying to figure out how to integrate these new tools into their administrative and instruction programs.

As an example, when a principal shares a slide deck that was generated with an AI tool, they should identify the tool and potentially how they used it so other staff can see the possibilities of AI tools and know when and how their school supports the use of AI tools.

• AI Resources for District Leaders
• AI Starter Kit for Teachers

I’ve been aware of Claude AI for a while now, but only recently sat down to test the AI’s latest free model Claude 3.7 Sonnet. This let me see how it stacks up against other AI tools such as ChatGPT, Gemini, and Grok.

After experimenting using Claude AI on various teaching exercises, from generating quizzes to tutoring and offering writing suggestions, I found that it's not perfect, but I think it behaves favorably compared to other AI tools in general, with an easy-to-use interface and some unique features that make it particularly suited for use in education.

Here is why Claude AI has become one of my go-to AI tools and one other teachers should consider adding to their AI toolkits.

1. Claude AI Prioritizes Privacy

Data security should always be a priority for educators, especially if they are encouraging students to use AI themselves.

One of Claude’s main selling points is privacy. Conversations users have with Claude are not used for training data, unless users opt to allow it. All conversations are also encrypted.

University educators also have the option of working with Claude for Education, which can provide additional security protections.

Of course, even with these privacy policies, teachers should make sure Claude or any AI tool is approved by their school IT and leadership teams before encouraging students to use it, or providing the platform with any kind of student data.

2. Claude AI is Less Likely to Hallucinate

Claude AI was developed by Anthropic, a company founded by siblings and former OpenAI members, Dario and Daniela Amodei, and responsible AI development is one of the company’s guiding principles. In keeping that and its approach to privacy, Claude AI is designed to prevent hallucinations with specific instructions to say, “I don’t know,” and to provide citations and transparency around how it reaches its answers.

I saw this firsthand when Claude refused to answer questions it didn’t know, such as who won a recent New York Mets game. This is important, as The New York Times reported that even while getting more powerful, many AI models are becoming more likely to hallucinate.

Though it performs better than some other AI tools I’ve used when it comes to hallucination, Claude is far from perfect in this regard, so its answers still require verification. For instance, when I asked it to help me create a slideshow for a recent talk I gave on vodka history and its surprising links to Bethel, Connecticut, it failed to find the actual connection — Smirnoff’s first U.S. distillery opened in Bethel in 1934 — and instead invented a local distillery called Asylum Distillery, even though no such spot exists.

That said, it did include a note urging to check if reference to this distillery was applicable, so there was a little bit of uncertainty there, which could be helpful.

3. Claude AI Is a Friendly Tutor

A complaint I have with many OpenAI products is that the interface can be a little clunky. Claude AI is designed to have a friendly conversational tone. This might have a subtle influence when it comes to tasks such as tutoring.

When I ask ChatGPT a question in the standard text version, it can feel like it is spitting back web results to me, whereas Claude often feels warmer. This could be key when using AI as a tutor, as a rapport with a tutor is so important for students.

Personally, I found Claude to be an engaging and accurate guide when I asked it to better explain to me how radio waves are actually a type of light wave. Another bonus, is that when I asked it to write a paper for me based upon a prompt I provided students, it tried to tutor me instead. Though on the negative side, it did provide a "sample" paper. It urged me to use it as a guide for writing my own paper, but its a stretch to think all students would take that advice to heart.

4. Claude AI Performs Admirably On Common AI Tasks

These days, many different AI tools do a lot of things very well. Therefore, deciding which ones to use can often be about the interface you like best or which companies better align with your values.

If Claude’s privacy and AI-safety-first approach resonates with you, from my tests asking it to generate simple quiz questions or project outlines, it performs as well and, in some cases, better than other AI tools for tasks. It can also help find errors in writing and caught several mistakes in samples I provided that other spell checkers I used missed — even pointing out repetitive word use. However, I wasn’t particularly impressed with its writing suggestions. The rewrites it suggested were, as is the case with most AI writing, clichéd and lifeless.

Despite this, overall Claude AI is a worthy competitor to other AI chatbots with some unique features that make it appealing to me as an educator.

• Claude AI: How To Use it To Teach
• AI Starter Kit for Teachers

In December 2024 Google announced its quantum computer chip Willow performed a computation in a few minutes that would have taken a classical computer more years than the age of the universe. Microsoft claims it invented a new state of matter for its quantum computers, and IBM has started offering quantum cloud computing services.

Lance Key, educator and support specialist at the Putnam County School System in Cookeville, Tennessee, says that teachers and the education field as a whole will likely see a significant impact from the new technology.

Here’s what educators should know about quantum computers and how these may soon influence a classroom near you.

1. Quantum Computers Use Qubits Instead of Bits

A classic computer uses bits: tiny switches that can be positioned “off” or “on” and signal the 0 and 1 of binary code.

Quantum computers use qubits, which are made from electrons, photons, or other particles. These tiny particles take advantage of the strange rules governing the quantum world, including a particle’s ability to exist in multiple places at once before it is observed. Because of this, qubits are often described as having the capability of being both 1 and 0 at the same time. Furthermore, these can take advantage of other strange rules of the quantum world, including quantum entanglement to operate in unison.

To visualize this, Key recommends imagining interconnected spheres. “If you think about connecting two circles, you really tangibly can only touch those two circles at one spot. But now, when we talk about touching a sphere with two entanglements, you can touch those many different ways really quick. So something that would have taken 8 bits, or maybe 24 bits in the past to run, it's now going to be like two qubits.”

2. Qubit Are More Powerful Than You Can Possibly Imagine

The exponential growth of this process can be hard to fathom. With a few hundred entangled qubits, we’d have more computing power than there are atoms in the observable universe.

While today the most powerful quantum computers being developed by tech giants such as Google, Microsoft, and IBM do have hundreds of qubits, these still have accuracy issues and regularly generate errors, so it's a struggle to outperform classical computers except on specialized tasks.

Once accuracy and stability is improved, many predict quantum computers will be able to perform tasks such as accurately model new drugs and the potential side effects without human trials. These will also, of course, have significant impacts on the field of education.

3. Quantum Computing Will Jumpstart AI Development

“We are not far from having quantum computing take off,” Key says. “And when quantum computing takes off, we're going to be able to train our AI models so much faster than we were able to in the past.”

If you ask any AI model a question now, you’ll often notice a lag as it “thinks” about how to respond to you, Lance says. Quantum computers could speed that process up by providing enhanced training for AI models.

While teachers and students won’t necessarily be accessing quantum computers, the AI models they use will be trained with the help of quantum computers operated by large tech companies.

“When quantum computing gets here, we're going to be 1,024 times faster than any computer that we have right now,” Key says. “AI on the front end is probably going to look the same, but AI is going to get a lot smarter.”

In that vein, more efficient quantum computing will help to stem the high energy needs and rapidly growing carbon footprint of current AI systems.

4. Quantum Computing Can Help Teachers Differentiate

Key believes quantum computers will help teachers to better assess the support students need by improving existing tech tools. These computers will do a better job looking at student data and recognizing the differentiation strategies each individual student might need, Key says. These can also vastly improve existing AI tutors.

“It is hard for me to replicate myself in the classroom to get around and answer every single question that comes out,” Key says. “But I think in the near future — I know there are people already working on it now — every student will have the ability to have an AI tutor.”

5. Quantum Computing Might Influence Your Classroom Sooner Than You Think

Quantum Computing, according to many people, is currently in a similar place to where the generative AI field was prior to the release of ChatGPT by OpenAI at the end of 2022. Large language models were getting better and better, and eventually made a major leap forward, but people didn't realize that until the general public got access to ChatGPT.

Key thinks quantum computing’s ChatGPT moment will come over the next few years. “Your Amazons, your Googles, your Microsofts, all of those companies are investing money because they see the future,” Key says. “They see what's fixing to come. I think in education, sometimes we're so ingrained in our work, we don't have time to see what's coming three to five years down the road.”

• Teaching Quantum Mechanics in K12
• 12 AI Tools Beyond ChatGPT That Can Save Teachers Time

This website uses Google machine learning smarts to allow people to play with the tech in a simple way that makes it highly accessible.

This can be used as a teaching tool to help educators get across how AI works. Students are able to actively explore the AI learning algorithms in a way that's fun but also illuminating.

Since this is a tool designed to be creatively used, it represents a vast array of options for teachers to work with to help students better understand AI.

This guide aims to explain what you need to know so you can see how Teachable Machine could be helpful to your class.

What is Teachable Machine?

Teachable Machine is a website-based tool that allows anyone to play with an advanced AI that can be trained and used.

The idea here is to take the algorithmic smarts of a learning AI and put that in the hands of anybody. As such, this platform is able to learn from images, sounds, and movements, and then respond to these in future as needed.

Teachable Machine is a way for students to see how AI learns and then responds. It can be played with creatively for fun, but also to be used as a discussion point around how AI works and, indeed, how it is all around us in the digital world in the form of algorithms.

While this is a helpful introduction to AI, it can also be used in more complex ways that allow more experienced students to work with this tool as part of creating more advanced digital projects.

How does Teachable Machine work?

Teachable Machine can be accessed directly from a web browser without the need to sign up or share any personal information. Educators and students can start using this, on most devices, right away.

The simplicity is apparent immediately with a "New Project" icon that can be tapped to get started. You are then able to record using either images, audio, or movements.

Select one option, for example Images, and you are then asked to record yourself, or whatever you are using, for the first category. You can then do the same for a second or more. Next you select the Learn icon and the AI will go to work processing what it sees. Finally, you can choose to download to work with the output locally or upload to use it online.

In either case you can see, live, the AI working as you change out image one and two, and see it recognizing the differences live on screen.

What are the best Teachable Machine features?

Teachable Machine is very simple to use and clearly shows how the AI has learned the differences between the inputs with which you trained it.

While that is fun in and of itself, the ability to take that output and use it -- perhaps in a coding project or app -- is a really powerful tool. That applies both as useful in adding AI to a project but also as a way to show students how this works.

In addition, there are some dedicated teaching resources created by MIT that can be downloaded and used to further the learning possible on this site. The AI + Ethics link takes you to a selection of resources in Google Docs that can be used to teach and explore AI.

One example here is a lesson that involves teaching algorithms as opinions in which students write an algorithm to detect the best peanut butter and jelly sandwich. They can then explore what made the AI pick one over another, and dive into exploring data sets and how opinions can affect outcomes.

The Dancing with AI tool is for playing with the visual movement detection and having it affect outcomes in fun and creative ways.

How much does Teachable Machine cost?

Teachable Machine is totally free to use and does not require you to sign in or give any personal details at all to get working with the tool.

While Google says the training is kept locally on your device, it's worth considering this may still not be enough for privacy rules at all schools.

Teachable Machine best tips and tricks

Start simple
Use the tool, as a class, to detect a simple difference so students get the idea of how this works, then let them play with it.

Get ethical
Use the exercises to explore the ethics of AI and the wider internet -- using YouTube as a great example of how algorithms work to filter our perspectives.

Export
For more advanced students, have them use the output as a way to create in their coding projects using advanced recognition techniques.

• New Teacher Starter Kit
• Best Tools for Teachers

For many students, knowing when to review which lectures or when to complete specific project components takes time, usually an hour or two to build a concrete plan of attack. However, that planning can be cut down significantly and optimized by using AI tools such as ChatGPT.

Currently, I’m preparing for my first board exam in medical school, a nine-hour test with 352 multiple-choice questions covering every foundational area of osteopathic medicine. To organize my six-week dedicated study period, I asked ChatGPT to help create a study schedule. I’ll outline here how I used it so educators and students alike can replicate or adapt the process for lessons, standardized test prep, or large-scale projects.

Planning matters, and procrastination kills progress. The method I describe here can help you quickly create more precise, robust, and realistic schedules.

AI Prompting: Start with the End in Mind

A tool is only as good as the person using it. The same goes for AI, and it all starts with effective prompting.

When I begin a new request, I first state my end goal and then ask ChatGPT what it needs from me to achieve that result. Here was my initial prompt:

“I need to make a study schedule for my 6-week dedicated study period starting on May 12th and ending on June 16th (the day of my COMLEX Level 1 exam). I want to include a lot of details. What would you need from me to help?”

ChatGPT responded by listing all the inputs it would need, effectively giving me a "worksheet" to complete before it could generate a proper schedule.

Completing the Worksheet

You’ll usually get a set of open-ended questions designed to clarify and customize your project. Here’s what ChatGPT asked me:

1. Daily Time Availability

• How many hours per day can you study during the week and on weekends?
• Do you have any mandatory obligations (appointments, research, etc.)?

2. Resources You Want to Use

• e.g., Pathoma, Sketchy, Boards & Beyond, Bootcamp, UWorld, COMBANK, TrueLearn, AnKing deck, First Aid, Savarese

3. Topics/Systems You're Weak In

• What do you want to prioritize?
• Any topics you’ve mastered and only need to review lightly?

4. Practice Exams

• Are you taking NBME, COMSAE, or COMLEX practice tests?
• What dates?

5. Study Preferences

• Prefer focused subject days or mixed reviews?
• Do you want built-in review or flex days?

6. Flashcards (Anki)

• How many new/review cards per day?
• Should Anki be scheduled into the plan?

7. Catch-up or Rest Days

• Do you want scheduled buffer days or a regular day off?

This step ensures the AI understands the nuances of your needs before generating anything.

Refine: Feedback Makes It Better

Once ChatGPT delivers a draft, usually a sample week/day when making a schedule, you’ll likely need to refine it. Even though these large language models are great at organization, the nuance that humans bring to personal planning can be missing.

For example, I disliked the date formatting ("2025-05-12") and requested a more readable style ("May 12"). I also noticed the AI scheduled three days for cardiovascular review when I only wanted two. This back-and-forth is crucial, just like giving feedback to a colleague. The more specific you are, the better the final product.

Finalize: Export and Customize

After refining a schedule, you can export it into a format that works best for you. ChatGPT can generate:

• A .csv file for Excel or Google Sheets
• A .pdf for printing
• An .ics file for syncing with your Google Calendar (my personal favorite)

These outputs turn your AI-generated plan into an actionable calendar you can further tweak.

Just remember: the AI provides a fantastic draft—you still need to tailor it. Add granularity where you need it, remove excess, and make it your own.

Conclusion

AI tools such as ChatGPT can be powerful assistants when it comes to scheduling individual projects and academic goals. Used properly, these can save hours of prep time and provide a strong framework to build on.

Whether you're creating lesson plans, study schedules, or managing larger projects, the process remains the same: Prompt, Provide, Refine, Finalize. After reading this article, I hope you feel more confident using ChatGPT not just as a Q&A tool but as a time-saving planning partner.

• 10 Ways Educators Can Use AI to Streamline Daily Life
• AI Resources for District Leaders

The tool, developed in-house by the district itself, aims to foster honesty and alleviate anxieties surrounding AI in education. It essentially works as a dropdown menu option in Google Docs that allows teachers to set parameters around AI use for specific assignments, and then lets students document which AI tools they used, and to what extent.

“I don't want my teachers to be spending time trying to catch kids cheating,” says LBUSD Chief Technology Officer Michael Morrison, who oversaw the two years of development for AI Trust You. “That's the opposite of why they got into teaching.” Having this tool helps teachers be clear about their expectations and students be clear about their AI usage, which reduces stress for both.

Since formally launching AI Trust You within LBUSD in January, the district has made it available for free to school systems across the nation. At last check, the extension is already being used by 47 districts and 600 teachers, although those numbers increase every week.

“We had one person from New Jersey who said that when she showed it to the principals this one high school principal got the chills,” says Morrison. “I think it's more that they realize they've been dealing with this issue with their teachers, and their teachers are frustrated, and it's a pain point for everyone. So I think we're addressing that in a unique way with this.”

Getting Started With AI Trust You

Anyone who is interested can go to the AI Trust You home page for a demo, and to sign up to get started.

The beauty of AI Trust You is its simplicity. Once installed, the extension is accessed via dropdown menu for any assignment through Google Docs.

For teachers, they start by selecting from the four AI technique categories—Content Creation, Content Understanding, Research Assistance, and Audio-Visual Generation—to specify how students are allowed to use AI on a particular assignment. The tool generates a clear assignment guideline text for students, so expectations are transparent.

From the student perspective, they select the AI tool or tools they use while working, and identify the techniques and the percentage of their work that was AI-assisted.

AI Trust You then generates a citation documenting exactly how AI was used for that assignment, which the student can then paste into their work, either at the end or wherever it’s required.

Success So Far

Although it’s only been in use a relatively short time, the tool seems to be working as intended, especially in regard to transparency.

“In one classroom, we had the teacher make all the kids use it, and then the teacher sat down with each kid and looked at the results,” says Morrison. “They went over what the kids said they used it for, where and how it was used in that assignment, and incredibly, it was right on accurate! The teacher said, ‘I could not believe how honest they were about how they used it.’ I think it proved to show that kids want to be honest, and given an opportunity, a lot will be. It's pretty cool to see.”

AI Trust You is not designed for every assignment, Morrison cautions, but for key ones in which teachers want to be clear about AI usage.

“I can picture teachers saying, ‘No, don't use AI at all,’ right? And that's appropriate at certain times,” he says. “I think teachers need to think about being more intentional about what they expect from kids with a specific assignment.” He suggests using AI for research isn’t all that different from using Google, so AI Trust You at least allows students to document the research process.

Improving School Culture

As mentioned, the extension itself is free, and will remain so, says Morrison. The district is working on providing a way to monitor stats on usage, for which it will charge to recoup development costs. To protect student privacy, no personal data is collected. Any information tracked is around the actual platforms used.

Ultimately, AI Trust You has lessened student apprehension about using AI, and reduced teacher suspicion around potential cheating–a big win-win as far as Morrison is concerned.

“I think this has the potential to make our school culture better,” he says. “And that's a pretty hard thing to accomplish with other tools, and this is so simple. I feel like it's almost a no-brainer.”

• 5 Ways Schools Can Prepare Students For AI
• 10 Free AI Trainings and Certifications for Educators

Sponsored by itopia, this webinar, which took place on April 8, 2025, explored how districts can take charge of their cybersecurity strategy, ensuring proactive threat prevention, responsible AI adoption, and total data control, with a platform such as itopia’s SecureClass. With fully customizable access controls, SecureCless helps you determine exactly what's allowed, down to the details.

Hosted by Tech & Learning’s Content and Brand Director Christine Weiser, the discussion featured Kyle Berger, Chief Technology Officer at Grapevine-Colleyville ISD, and Jena Draper, Chief Innovation Officer at Itopia.

Watch the replay here

Key Takeaways

AI Adoption and Cybersecurity Risks: AI is rapidly being adopted in education, far exceeding that of previous technologies such as the internet or social media, however, this growth brings increased cybersecurity risks. Schools need to secure data from AI-driven threats and maintain control over their systems. “Tools have now embedded AI into their platform, sometimes with us knowing it and accepting those terms, and sometimes without us accepting or knowing those terms,” said Draper. “It has created this extreme risk for us." This necessitates continuous review and adaptation of cybersecurity strategies.

Balancing AI Adoption and Security: "AI is going to be embedded in jobs that are existing today,” said Berger. “Jobs that have always been around are adapting." Consequently, schools cannot simply block AI due to its increasing importance in preparing students for the future workforce. Instead, they must find ways to enable AI use while mitigating risks.

Threat Landscape: The threat landscape is evolving, with bad actors increasingly using AI for malicious purposes. Schools need tools and strategies to defend against these advanced threats.

Data Control: A significant focus for schools and AI should be data control. “It just comes down to what data can be shared with that AI tool, right?” said Draper. “If I'm just typing in silly things about cats or something, it doesn't really matter. But if I'm starting to put my personal information in there, whether it's knowingly or unknowingly, then that's what creates the risk.” To that extent, Itopia's SecureClass allows schools to control what data can be shared with AI tools, such as blocking copy/paste, uploads/downloads, and camera/microphone access.

Chromebook Security: A misconception exists that Chromebooks are inherently secure. However, as their use increases, so does the risk. “It's not just students that are using these devices anymore,” said Berger. “The more usage you see, the more targets and threat actors are gonna start trying to find ways to get to it.” Ensuring that your cybersecurity protection works for all devices is critical.

Fighting AI with AI: The webinar emphasized the need to use AI to fight AI threats. “We've got a massive database of over 8 billion endpoints, and we're watching these threats as they are suspicious events so that they don't become problematic for you in the future,” said Draper. “So anytime a URL has any kind of suspicious activity, we go ahead and block it with our extension in real time, and we unblock it proactively as we get information that allows us to feel that that is no longer a risk to you. So we've blocked over 18 million websites in just the last 90 days for suspicious activity around malware and phishing. This is no easy task. But securing AI and then using AI to secure it is all part of what SecureClass's underlying mission is.”

• Tech & Learning Webinars

As with any AI trend of this magnitude, OpenAI’s new ChatGPT-linked image generator is bound to make its way into the classroom whether teachers want it to or not, so I recently began exploring some of its features in general, and its potential for teaching, overall.

My verdict is that it's the most impressive AI image-generating tool I’ve worked with and has a good deal of teaching potential, particularly when it comes to classroom prep. However, its knowledge of biology left something to be desired, and it brings with it a hornet's nest of ethical questions and concerns.

What OpenAI’S New Art Generating Tool Does Well

The new art-generating capabilities from OpenAI’s GPT-4o model create better-looking and more natural and realistic-feeling images than other AI generators I've tried, even when it’s creating fantasy scenes. For instance, the image below of a dragon flying over New York, though not spectacular, is better than images I’ve created with other AI art generators.

My prompt: Can you create an image of a dragon flying over New York City for me? The dragon should look like a classic fantasy dragon but the city should be realistic looking.

I also was able to use the tool to generate some potential writing class materials for teaching students to create thesis statements. Though getting what I actually wanted probably took more time, if less mental effort, than creating an illustration on my own would have.

My prompt: You are a college professor teaching an introductory course in English composition to freshman students. Can you create a diagram that visually illustrates how to write a thesis statement in the Ghibli Studio style?

My follow-up prompt: Please create a new image that gives examples of an arguable thesis and one that does not. This example should argue for or against the benefits of AI in education.

This is a good overview, but I thought the example of an arguable thesis could be strengthened by providing more reasons, and that the whole illustration could use a visual upgrade. Getting the AI to do this took a few tries.

Finally, the AI produced this:

Where OpenAI's New Art Generator Fell Short With Teaching

Even though I was ultimately pleased with the example generated above, it took me a while to get there. As mentioned, it would have been quicker for me to just type out a few examples on my own and paste these against a generic slideshow backdrop.

When I first started teaching and would constantly second-guess myself, a tool such as this probably would have saved time. But I do wonder if the process of creating examples on my own helped me clarify my thoughts on the topics I teach and develop as a teacher, which might've been lost by using AI.

More troubling was my attempt to have the AI create a diagram of an animal cell.

My prompt: Can you create an annotated illustration of an animal cell for a middle school biology class?

As Sam Cooke put it, "I don't know much biology," but I'm pretty sure most cells have one nucleus, and it's, like, located closer to the center, maybe?

Teachers, however, can guard against these types of mistakes. The worst thing about OpenAI's new art-generating capabilities is it continues the trend of large tech companies profiting off of art they did not create originally without compensating artists. Artists at Studio Ghibli, for instance, deserve compensation for all the images being created in the style of their distinctive and iconic work.

Bottom Line

OpenAI's new art generator is one of the best around, and even though it's not particularly helpful for my teaching purposes, I'm sure other teachers who are more creative than me will develop innovative ways to use to inspire students. But with these new abilities come new ethical questions. Right now, those ethical implications are too murky for me to use it to create class materials.

That said, it is more fun than any AI tool I've used recently. Like, addictively fun!

Here's it taking my headshot and giving it the Studio Ghibli treatment.

And here's what happens when I asked it to give me a sword and send me on a quest. My misgivings about this technology notwithstanding, I will admit my Studio Ghibli avatar is looking sharp and appreciate that he seems to spend more time at the gym than I do.

• ChatGPT Talks Now. What Does That Mean For Teaching?
• GPT-4.5: What Educators Need To Know About The New ChatGPT Mode

Fortunately, artificial intelligence (AI) can now act as a personal assistant, helping you to save time, reduce decision fatigue, and restore some work-life balance.

Here are ten practical ways AI tools can streamline your daily life outside the classroom:

1. Create Smart Grocery Lists

AI tools such as ChatGPT or AI-enabled apps (e.g., AnyList, Google Keep with AI integrations [still in Beta]) can generate grocery lists based on your meal plan, dietary preferences, or even what’s already in your pantry. For example: Ask ChatGPT: “Make a grocery list for five easy dinners with chicken, beans, and rice as main ingredients.” You’ll get a ready-to-shop list that is customizable and shareable in seconds. You could similarly identify the ingredients you currently have and ask AI to identify potential meal options.

2. Plan Weekly Meals in Seconds

Decision fatigue is real, especially after a long day at school. Let AI take over weekly meal planning. You can include time limits, dietary needs, or ingredient preferences. For example: “Plan five quick vegetarian dinners I can make in under 30 minutes, with minimal prep.” You can pair this with a grocery list generator to get everything in one go. Additionally, if you forget an ingredient, AI can provide a list of potential substitutes based on what you have available.

3. Summarize Long Emails or Articles

If you’re subscribed to parent newsletters, community updates, or other info-heavy content, AI can summarize these for you in a flash. Browser extensions such as Wordtune Read, or plugins for ChatGPT or Chrome, can help you focus only on the key points. As an example: Paste an article into an AI tool and ask: “Summarize this in three bullet points.” For more robust summarizing tools, see 8 AI Tools for Summarizing and Reviewing.

4. Auto-Draft Thank You Notes or Messages

Whether you’re thanking a friend for dinner or following up with a neighbor, AI can help draft warm, personalized messages without the stress of figuring out the right words. For example: “Write a friendly thank-you note to a neighbor who brought over cookies.” Revise the output as needed, and it can save time and mental energy.

5. Schedule Social Media Posts or Reminders

AI can help you brainstorm captions and post ideas for birthdays, holidays, or even school events. Some tools can help automate reminders and calendar tasks using natural language. An example could be “create a cheerful Instagram post for my son’s 10th birthday with a short poem” or “create an image to celebrate my husband’s (our) 10^th wedding anniversary.”

6. Organize Travel Plans

Planning a weekend trip? AI tools can act as mini travel agents, suggesting destinations, building itineraries, and even helping you find the best travel times or routes. Try “Plan a three day weekend trip near Minneapolis with historical sites and budget lodging” as a sample or “build a day long road trip route through Minnesota that visits points of interest including the World’s Largest Ball of Twine and the SPAM Museum.” Each state has similarly interesting roadside attractions.

7. Create a Study Plan for Your Own Coursework

In order to make your own classwork fit more effectively into your schedule, ask Google Gemini or another tool to review your syllabus and ask it to draft a study schedule to ensure all your assignments are submitted at least a day before the due date. You could similarly have AI draft your spring workout schedule to prepare for that mid-summer half marathon!

8. Draft and Revise Household To-Do Lists

Let AI help create or prioritize to-do lists for house chores, errands, or home projects. You can ask it to organize by urgency or time required, or even assign tasks to other family members. Try using “create a weekend to-do list for cleaning, grocery shopping, and minor home repairs.”

9. Plan Family Activities or Game Nights

AI can suggest creative ideas for rainy-day activities, family movie nights, or screen-free games for kids of various ages. Claude could be asked “suggest fun, low-cost family activities for a Saturday afternoon with two kids aged 8 and 11.”

10. Reflect and Journal with AI

End your day with AI-assisted journaling. Tools such as ChatGPT can prompt you with questions to help you reflect, practice gratitude, or simply de-stress. “Please give me a journal prompt to reflect on a challenging but rewarding moment today.” It is potentially a low-effort way to check in with yourself, reflect, and stay grounded.

Living Smarter

Educators are caregivers, knowledge-sharers, and multitaskers, and it’s easy to put your own needs on the back burner. By offloading some of the daily mental tasks to AI, you’re saving time and freeing up brainpower for the things that matter most: being present with your students, your family, and yourself.

The key is to start small; try one or two of these ideas this week. You’ll likely be surprised by how quickly AI becomes your time-saving sidekick.

• 9 Suggestions For Your AI Reading List
• AI Resources for District Leaders

Megan Simonson

Megan Simonson earned a bachelor’s degree in psychology from Saint Mary's University and a master's degree in School Counseling and Community Counseling from Winona State University (WSU). She has worked in various enrollment-related roles in WSU’s College of Education since joining the team in 2016 while also serving part-time as a mental health counselor at an outpatient treatment center in Rochester, MN. She now focuses on her role as the Recruitment and Admission Coordinator for the College of Education.

I started teaching back in 1995—long before AI became the household acronym that it is today—after earning my engineering degree and a master’s degree in technology. I learned about VHS Learning through a College Board forum about five years ago. Since joining the company, I’ve revised and expanded our computer science program to provide critical opportunities for computer science education to students around the world. I’ve developed courses including AP Computer Science A, AP Computer Science Principles, Discrete Mathematics, Cybersecurity, Data Science, and our newest course piloting this spring, Artificial Intelligence.

5 Ways Schools Can Prepare Students for AI

Artificial intelligence has been around for decades, but now that it is an integral part of all our lives, we must support our students in developing the skills to be informed users of it.

At VHS Learning, our target audience for our Artificial Intelligence course is the high school student who is curious about AI and the impact it has on their life. In our work to develop this exciting new course we identified five important goals that may be helpful for any school that is considering how to teach students about AI:

1. Expose students to a wide variety of applications of AI: While most students are aware of AI, they may have a very limited knowledge based on fictional representations and social media influence. In our AI course, students are constantly exploring the use of AI across many different fields. For example, students may not realize that AI-driven medical cancer screening programs—approved by the FDA a couple years ago—can now detect cancer earlier because the programs are trained to detect abnormalities beyond what the human eye can detect. Prompting students to explore how AI is used in various fields and contexts is a great way to get them interested in learning more about how artificial intelligence works.

2. Present the big picture of AI: Most students assume a technology degree will in some way involve computer coding, but do not understand how AI is truly the intersection of programming, machine learning, neural networks, and deep learning. Students do not need extensive programming experience to have a basic understanding of how AI works. In our course, for example, students can explore how AI is based on computer code without any prior programming knowledge. They are provided with friendly versions of code that allows an app to function and are then prompted to explore how making changes in the code affects the functionality of the app. Students then shift their focus to explore the general concepts of machine learning, neural networks and deep learning in later units of the course.

3. Provide opportunities for hands-on experience: In the beginning of our AI course we use Google’s Teachable Machine to allow students to create an AI model that is trained to distinguish between two animals. They upload photos of their favorite animals and then input data to train their model to identify the different types of animals by key characteristics. Students explore how processing more images with their model allows the AI app to get smarter by helping to identify where further training is needed. This is a very simple yet effective model to teach students the power of AI, how machine learning works and why limitations or biases may exist.

4. Explore the impact of AI on careers: We ask our students to try and identify an industry that’s not going to be impacted by AI in some way. As high school students consider their future and potential career fields, we want to ensure they consider how AI will impact those fields. This is yet another reason why schools should be embedding AI learning into their curricula across disciplines. No matter what career students choose, they will need to have a general understanding of AI, how it works, and how it’s being used -- as well as how it shouldn’t be used.

5. Engage students in critical thinking about the ethical implications of AI: We recognize that AI can be a scary topic for students and set out to make the ethics of AI a central theme in our course. Each week students discuss applications of it and consider both the benefits and potential harm of those applications. In our final module, Living in an AI World, students spend time exploring human centered design and the responsible use of AI, including ethical frameworks, privacy and empathy maps, and consider the future of AI. When developing this course, we were intentional about looking at AI through different lenses, to allow students to understand the complex nature of the technology and its applications, to empower them to make educated decisions about their use of AI, and to be informed decision makers that will help shape future AI policies.

The Responsibility and Challenge of Teaching Students about AI

Instead of shying away from AI tools, we should be integrating them into curriculum from day one. Students need to learn not just how to use these tools, but also how to critically evaluate their output. We've seen students attempt to use AI to cheat, demonstrating a troubling lack of understanding of AI’s limitations, such as its propensity to "hallucinate" or generate inaccurate information. With school-appropriate AI platforms, we can teach students how to use AI responsibly, analyze generated content, and delve into the source of the information to investigate limitations and bias—all while preparing them for success in a wide variety of careers and industries.

Finally, there’s a perception that today’s students are very highly technically trained -- this is simply not the case for many. While there is good progress being made in expanding access to computer science education, the bottom line is that many schools still don’t have access to qualified teachers who can teach subjects such as AI. By using a high-quality online learning program such as VHS Learning, which features many highly qualified computer science teachers who bring a wealth of knowledge, experience, and passion to our program, schools can equip their students with the essential knowledge and understanding of technological advancements shaping their future, regardless of their prior exposure to computer science.

Unlike other AI services out there, this is designed specifically for education and -- even more targeted -- to help educators and kids in certain areas of need.

So, essentially, this is a chatbot being called an education assistant. But what gives this so much weight is that it's been in development since 2021. That's meant interviewing and surveying more than 500 teachers before co-designing, again with teachers onboard throughout the process. For math assistance, math teachers were used specifically, for example. All this the company calls its human-centered AI partnership model.

This guide lays out all you need to know about Colleague AI to see if this could help in your class.

What is Colleague AI?

Colleague AI is a chatbot, broken down into task-specific chatbots, designed and built to support education. This works with teachers, students, admin staff, and parents, all as an assistant.

The chatbots are designed and vetted by teachers, federal agencies, educational scholars, and AI experts to be FERPA and COPPA compliant, and get an end result that's as safe and useful as possible.

This spans from assisting teachers with planning lessons to working on professional development. It also works with students to help with homework, track learning progress, personalize learning, and more.

For school leaders, this can provide data-driven insights and teacher support systems as well as automating administrative tasks and helping with stakeholder communications.

Finally, parents are also supported with a chatbot that can help create learning tools for outside of school, and even allows parents to grade and connect with other kids to see progress.

How does Colleague AI work?

Colleague AI can be used as a native software via browsers or as a Chrome extension. In either case, teachers are able to work with the AI characters of Charlie or Claire, to lesson plan, mark work, and more.

Usefully, this allows you to go back and forth in a conversational way. So a teacher might pick a lesson from material that's already available, then prompt the AI to change that to teach a certain area, in a certain timeframe, targeting specific students. It will then offer an adjusted output and, again, you can ask for more changes until it's just right.

While it can be open-ended with written statements to get results, there are also helpful drop-downs along the way that provide options to vary certain areas as needed. This can be helpful to get an idea of what is possible or to get quick variations on what you're working on at the moment.

What are the best Colleague AI features?

Colleague AI is a great way to help with lesson planning as it allows teachers to pull-in premade free-to-use lesson plans, which can then be augmented. It is also possible to start from scratch or upload materials to manipulate.

This is really helpful at reading handwriting, so students can upload photos of their work and you will be able to have the AI pull out the text and mark it automatically. This is graded in sub-sections for helpful breakdowns, and allows you to adjust aspects such as rubrics or outputs, as needed.

You can adjust by selecting subject, grade, and even state standards to get the overall output you want. Then, in a chatty way, go back and forth with the AI until it's just what you need.

A great thing about how easy this is to output and mark is that it can offer really learner-centered lessons and feedback, all without taking up more teacher time. In fact, the makers say this can save teachers up to 50% of their time.

The Chrome extension is a useful option as it takes the AI with you into the browser, allowing you to adjust, edit, and draw any information you want -- a great way to integrate Google Drive marking, for example.

How much does Colleague AI cost?

Colleague AI doesn't appear to mention anything about pricing on its website so it seems everything, at time of publishing at least, is free to use.

This is often the case with AI tools, as the company can improve output based on data collected from human use. So whether this stays the same longer term is not clear at this stage.

Colleague AI best tips and tricks

Start simple
Use the lessons already available and edit these using suggestions only to get a feel for what's possible with the AI.

Mark written digitally
Use the Chrome extension to mark handwritten homework that's submitted via pictures and a digital drive, automatically.

Stay organized
Use the online storage space to keep lesson plans and rubrics you've created for future use, or to edit for personalized learning variants.

• New Teacher Starter Kit
• Best Tools for Teachers

Pete Just, Executive Director on the Indiana CTO Council, discusses the intricacies of approaching AI in education, AI adoption strategies, and how to assuage the fear that comes with AI use in the classroom. Just recently presented on this topic at the spring CoSN conference.

The Adoption Curve for AI

When tech innovation is added to education, the adoption process can be swift or it can be measured. AI, as amazing as it can be at times, seems to be taking more of a measured trajectory in schools.

“I thought we would follow a much more aggressive adoption curve because of the transformative nature of what generative AI brings to the table,” says Just. "But it also brings concern, and we need to mitigate that concern. What I mean by the general adoption cycle is that everyone is waiting for the other person to make the change and then they watch what they're doing and they say, ‘Okay, that's not too bad. That's got some benefits. We might want to do that.’”

Many are waiting for the “right” way to use AI in the classroom to present itself. Whereas some are willing to take the leap into AI use, many are hesitant when it comes to altering how we teach students. One of the biggest challenges is shifting this mindset so that students can gain practical experience using AI as it will undoubtedly be a significant part of their future careers.

Changing The “Using AI Is Cheating” Narrative

If a student uses AI for an assignment, is it technically a form of cheating? Can it be labeled as plagiarism? Does it hurt a student’s ability to think critically or problem solve?

”If you are not helping your students understand what we're talking about with generative AI, what it could bring, how to use utilize it, how to think critically about asking those questions or prompting, how to think critically about reviewing the answers or the output, you are doing your students a disservice,” says Just.

What about those who are using AI in the classroom effectively?

“They don't see it as cheating. They see it as amplifying,” says Just. “And so what we have to do is we have to get ahead of that. What we're doing is helping school districts think strategically about how they can amplify what they do within their school district and really turbocharge it in a lot of ways, while also protecting student data and putting guard rails in place.”

AI has become a tool that can help bring students closer together in terms of academic achievement. Crafting ideas, bridging understanding gaps, and enhancing participation can all be supplemented with the addition of AI, as long as teachers demonstrate how to do so effectively and ethically.

The Role Of Administrators in AI Adoption

Should an administrator using AI in their personal life promote its use in a school or district regardless of existing guidelines? What exactly should administrators do when it comes to AI use in education?

“There’s a group of people who are in charge,” says Just. “And their way of taking charge is by having the conversations with their school board, by having the conversations with their community stakeholders, by having the conversations with their teachers, and then determining what is the feeling in our school community about where we should go with this.”

Many education decisions can be held up due to unwillingness to act, so keeping open lines of communication with stakeholders regarding AI can help guide the conversations that are needed. As Just puts it, AI can do amazing things for teachers, but they need the support of their decision-makers to get where they want to go.

• Explaining Your School’s AI Approach To Parents
• Using Chatbots To Create Personalized Learning Experiences

I was glad I went to that session on how to mail merge two decades ago. I learned how I could economize time by simply making a message that was more personalized and sharing it uniquely with various recipients. This was helpful with a full class, grade, club, or even schoolwide notification. The personalization of mail merge helped me uniquely identify recipients by their name and specifics related to them, such as their child, the relevant grade, perhaps a specific award, and other identifying features.

I have used the Google extension Autocrat to merge letters and found it useful, even if at times it can be glitchy and cumbersome.

The potential to give up on mail merge and just write each individually might frustrate those attempting to economize on a busy educator’s schedule. The problem is volume (if you have to send dozens, hundreds, or even thousands of messages) and an increase in errors.

Errors happen when you start rewriting the same general message over and over again, looking for those keywords to change, such as name, department or grade, and any other uniquely identifiable features, and your eyes start to glaze over. This is why mail merge can be more effective, content can be better coded and delineated in a spreadsheet that spools to a nice, clean mail merge.

Not to downplay, mail merge is still a great tool for those mass unique needs such as important recognitions or notices, and Autocrat has gotten me out of a lot of the heavy lifting of volume work, just as Microsoft has for others.

But what if there was an even quicker way to mail merge?

Enter AI.

Putting ‘AI’ In mAIl Merge

The other night I was at parent-teacher conferences and was scheduled to be out the next day, during which a letter had to be distributed to a specialized group of Career Day professionals. (Full disclosure, I had delayed preparing these and was in a hurry, and that’s when errors typically increase!) This group was listed in the school counselor’s spreadsheet who shared it with me, and contained the following basic information:

You might think: Okay, that’s not a lot of information. Big deal, you can just craft each letter.

Yet here’s where I went beyond mail merge, and added value to the letter. I attached the spreadsheet, in PDF format (simply take the Google sheet or spreadsheet, and select file, then save as PDF).

For this, I used Gemini. Other AIs of preference should work as sufficiently well.

Here was the initial prompt:

Use this text to mail merge the following volunteers to identify them by name and their profession, thanking them personally: March 14, 2024

What this spewed out was fine, but hard to track from one to the next, so I refined my prompt:

Put these into steps with numbering.

Now I could easily see the clear delineation from one letter to the next in the series and not be confused by free flowing text blending it all together. But it gets better! For each name, the AI identified a gender (“Dear Mrs. Zion…”). These were 100% accurate upon checking and assigned a title. For instance, one individual was a professor at Rutgers. The greeting was, “Dear Professor….”

Additionally, the AI cohesively entered the profession when appropriate.

Then I pasted each of the responses into a letter (works with email too). Having experience with mail merge previously, this finished my chore before conferences had ended that night!

I can safely say that this process of engaging the AI took less than half the time mail merge would because you don’t have to set any parameters linking the spreadsheet and document. The AI does that for you, and everything was grammatically correct, the appropriate information was identified, and it all made for a cohesive sequence of letters.

The Benefits of Using Mail Merge with AI

Ironically, I find myself using mail merge more frequently again, feeling the safety that the errors I accidentally entered bleary-eyed by not using mail merge are avoided. Also, AI can correctly identify individuals and other key information. And with more complex spreadsheets, I just tell it which columns to pull from, and AI does the work.

Time saved, clean, and personalized–well worth the nostalgia of mail merge, adding AI’s skills!

• ChatGPT Talks Now. What Does That Mean For Teaching?
• Gemini: Teaching With Google’s AI Chatbot

Laura Deiss, Curriculum and Instruction Specialist for Region 3 in Victoria, Texas, and Molly Kresta, teacher/instructional tech/grant writer for Louise ISD in Texas, discuss ways to implement AI synergy in your educational work space in order to achieve solutions for complex creative issues.

What AI Synergy Looks Like

If you’ve worked in a corporate environment, you might have heard of “synergy,” which is generally used to describe an activity that engages multiple parts to create a well working collective whole. This can mean people or technology, so long as all the separate parts work together.

In this case, if you remove Janet from Accounting and put in MagicSchool, you have AI synergy.

“AI makes [things] a little easier,” says Deiss. “So now let's start thinking about how to leverage it.”

Why isn’t AI tool synergy more prominent in the classroom?

“You think people will figure it out on their own but in the classroom, you just get tunnel vision, and so lots of teachers don’t think too much [about other tools they could be using],” says Kresta. “What I wanted to show [teachers] was there is not a one-trick pony with the AI stuff.”

When it comes to using AI, thinking outside the box can help teachers better understand how to combine the use of certain tools to create a favorable outcome.

How to Achieve AI Synergy

While it might sound easy in concept, a teacher might not be able to properly focus on multiple technologies simultaneously. Given the constraints and pressures that accompany the profession, they might stick to whatever tools they happen to use to get whatever solutions they can. However, a shift in mindset can help achieve AI synergy.

As Kresta says, putting the time into training teachers on AI tool use will pay off in the end.

“When you can find the time, when you can sit down and talk to [teachers] and show them [AI tools], then they’re more open to it,” says Kresta. “But you really have to sit down and talk to them. Teachers are too busy. You’ve got to find time to actually sit down and say ‘Here’s an idea.’”

So what would a good example of AI synergy look like?

“One of our science teachers came down and was asking me something completely unrelated,” says Kresta. “But he was using MagicSchool to do some things. I said, ‘After you did that, what if you did this? If you smash [different AI tools] together, take all these things I’ve taught you and put them together…’ You can save a teacher time and get students more engaged.”

In another example provided by Kresta, AI synergy was used to help transform a lesson on MLA into a multilingual experience and an interactive slideshow.

“For this lesson,” says Kresta, “I found a website with a great overview of MLA Style. I copied this into Gamma and it created the presentation (MLA Citations A Comprehensive Guide). I then downloaded the presentation as a .pdf and then used Google Translate to make a Spanish translation to distribute to my ELLs (MLA Citations A Comprehensive Guide Spanish). I also printed out copies for my students who require copies of notes as an accommodation. Once this was complete, I opened Curipod and used the Curify My Slides feature to add interactive slides between concepts (MLA Basics). This took 15 minutes to complete.”

The act of packaging AI tools together to solve a problem can vastly improve efficiency and reduce stress.

How to Get Over the AI Hump

Many are still struggling with the question of AI use in school, with some teachers on the fence about using it themselves. The first step of getting teachers on board with AI is finding out who is willing to take the plunge.

“It depends on who you talk to,” says Kresta. “I have a teacher across the halfway from me who is the most amazing teacher. I cannot get them to turn on a computer in their classroom except to put grades in. That’s it. So it depends on who you’re talking about and what they teach for the most part.”

Deiss says that it's also important to remind teachers that integrating AI into class will better help students prepare for their professional careers.

“We were at a conference where a Google exec said that they’re predicting that AI will be a constraint in jobs in the future,” says Deiss. “If you can’t use AI effectively, then that’s going to be a mark against you in the job market of the future.”

• Using Chatbots To Create Personalized Learning Experiences
• Explaining Your School’s AI Approach To Parents

Advance Voice Mode is a new feature that allows users to chat with ChatGPT in real-time. If you’re thinking that sounds like Amazon or Apple’s well-known voice assistants Alexa or Siri, you’d be right, but also wrong. This technology feels light years ahead of those often comically clumsy voice assistants.

Speaking with ChatGPT in Advanced Voice Mode is natural, quick, and so lifelike it is downright spooky at times. The process can feel as intriguing and unsettling as first interacting with ChatGPT did. It also has a feature that lets it use your phone’s camera to comment on your environment.

As an educator, I’m particularly interested in exploring the applications of ChatGPT’s voice assistant for education. To that end, I asked it to tutor me in various topics, and tested its ability to recognize features of the real world. Across the board, I was left impressed and expect this type of technology will become a part of some educational settings going forward.

What Is ChatGPT’s Advanced Voice Mode?

ChatGPT’s advanced voice mode is a feature that lets you talk with the AI chatbot rather than message with it. Users get to choose from several life-like voices of various genders, accents, and styles. I chose a voice with a British accent because it reminded me of Iron Man’s AI pal J.A.R.V.I.S.

Once you’ve selected a voice option, you can start chatting away the same way you might with the text version of the app. Though the experience feels very different. I found myself apologizing when I had to end a session, and feeling like I was on the phone with a real live person.

The merging of voice assistant technology and the latest generative AI models is starting to heat up. Amazon is readying the release of Alexa+ an LLM-powered version of Alexa. Sesame, an AI voice assistant app from Oculus co-founder Brendan Iribe, has gone viral and received great reviews. Perhaps to keep pace, OpenAI recently released updates for its Advanced Voice Mode, allowing for more natural conversations.

What Is Using This Tool Like?

To use ChatGPT’s voice assistant, all you have to do is open the ChatGPT app on your phone or navigate to the tool on your browser. On the right side of the text box is an icon with wavy lines. Click on that to enter voice mode and get ready for a different experience.

Speaking with ChatGPT’s voice assistant brought me back to the first time I interacted with an LLM: it was impressive and seemed like a technological magic show. But it was also a little disturbing. The tool generally responds in near real-time and with emotional-sounding inflections that sent my uncanny valley radar into overdrive.

Equally as unsettling is when I used it with my phone and permitted it to access my camera. When I asked it to tell me about my house, after panning the living room, it felt positively Sherlockian in its deduction skills. It could tell that I was likely a parent because of all the toys and correctly identified my dog as a labradoodle.

How Is It For Teaching?

The potential of a tool such as ChatGPT's voice assistant for education is pretty obvious.

One drawback to a traditional AI chatbot, however, is that the text interface can be cold and clunky (and yes, I just used the word “traditional” when talking about older generations of AI).

In theory, ChatGPT’s voice assistant should provide more lifelike assistance and could better work with younger children. Of course, questions around how young students will react to the life-like technology will need to be researched.

In my tests with it, I didn’t find it more helpful as a tutor than standard ChatGPT. Though I liked the idea of just asking a question in theory, in practice, having it explain some advanced physics equations to me was less effective than the text-based version because I couldn’t see what the tool was talking about.

Though ChatGPT's Advanced Voice Mode saves a text version of each conversation by default, it doesn’t refer to images in the same way a good teacher would talk and provide visual examples. This was also true when I tried to have it offer writing advice. It provided the same, sometimes helpful, sometimes generic writing advice you’ll find on the text-based version of ChatGPT. Switching back and forth between the text version of the tool and the voice version, however, made the process more cumbersome than just using the text version of ChatGPT alone would have been.

That said, when education-focused companies start pairing this technology with specific education tools, it has the potential to be very powerful. This type of tool also has high potential for use as assistive technology. For example, eventually it could read text to a student with visual impairment and describe the layout of a webpage or even a physical space.

ChatGPT Voice Assistant: The Bottom Line

ChatGPT’s voice assistant, and this type of technology overall, is an important part of the ongoing AI revolution in education and beyond with great potential for accessibility. AI-savvy teachers will want to be aware of this technology and might even want to start trying it out on their own.

Though it’s too soon to start encouraging students to use it — for one thing, I worry younger students might actually be scared of it — it’s almost certainly going to become a part of various educational tools in the not-too-distant future.

• GPT-4.5: What Educators Need To Know About The New ChatGPT Mode
• Google’s New AI Tutor LearnLM Is Trained On Learning Science and It Shows

Usefully, this does take AI smarts in order to save teachers on admin time. But crucially, it does this with standards alignment, so it can be very specifically targeted to learning goals and outcomes.

At time of publishing, this platform is being used by more than 5,000 schools worldwide. Helpfully, it integrates with a whole host of digital tools, platforms, and LMS options for the most simple implementation into a teacher's current setup.

This guide aims to lay out all you need to know about Almanack to see how it could work for your class.

What is Almanack?

Almanack is an AI assistant, of sorts, designed for educators. Consequently, this is tailored to very specific tasks and works to help teachers, just by being prompted.

This can create resources such as slide decks, worksheets, lesson plans, or report cards, and plenty more. Since this works across subjects, it can be used by teachers throughout the school to suit their needs.

The tools are designed for teachers, so students will only see the output content, if at all. The AI output doesn't go directly to them but rather works to help teachers have the resources they need when they want.

And since this is all created using written commands, it's simple to pick up and start using, for anyone.

How does Almanack work?

Almanack can be signed up to for free with many of the basic tools available. This immediately allows teachers to start lesson planning, differentiating instruction, and personalizing learning with the resources.

Integration is a big part of this with support to pull in and output using Google Forms, Kahoot!, Gimkit, Blooket, Quizizz, Quizlet, Google Slides, Google Docs, PowerPoint, Word, PDFs, and more.

Slide deck creation is a good place to start as that is one of the tool's most powerful offerings. Using written prompts, teachers can create anything they need to suit a task. That can mean a deck about a historical event, one about the year's plan ahead, a presentation on a new topic in science -- anything you can think of, really.

Occasionally, there can be odd output issues that need fixing, but even that works as a far faster option than traditional methods. For example, you can create multiple versions of a deck, with each one tailored to certain student groups, in as much time as it would have taken to make one for the whole class using traditional methods.

What are the best Almanack features?

Almanack is an AI, so like most tools right now, it's a starting point. A simple output might need some adjusting to get perfect before use in class. But, crucially, that's as much the human prompt issue as the AI.

Over time, as you improve in knowing how to prompt specifically for what you want, so too will the output improve. Ultimately, you will save more and more time as you get proficient at using the tool.

The ability to translate most outputs is also really helpful for multiple language versions and working with ELL and ELA students. You can have one for everyone in the class, using their native tongue, without it costing you much on time. And, perhaps most importantly, it's very accurate for a natural output that will be understood clearly.

The fact these tools are built for teachers means it's all standards-aligned, private and safe when it comes to data, and tailored to be viewed by children, so the content is going to be appropriate.

Everything still needs a human check first -- AI isn't perfect yet -- but it's a big step in quality and a huge potential saving on time for teachers.

How much does Almanack cost?

Almanack has a free version as well as several tiers of premium versions that add more features and tools.

The Free version gets you state standards/curriculum Integration, curriculum design and long range planner, lesson and unit planner, eduslides/presentations, worksheets/student activities, quizzes and tests, games, and custom video recommendations.

Go for the Pro tier, at $7.50/mo when paid annually at $89.99, and you get all the above plus Import to enhance with AI or build from scratch, a variety of ready to use worksheets and activities, differentiation based on DOK, exports, custom report card writer, integrations, and a limited number of collaborators.

Or for Small Teams, at $6/mo paid annually at $71.99, you get the Pro features plus an unlimited number of collaborators and team collaboration features.

Almanack best tips and tricks

Plan the year
Use the slide deck creator to plan your lessons for the year and output it so students can see what's ahead, as needed.

Tailor everything
Use the AI to help differentiate outputs so you have versions for ability levels to suit the entire class.

Translate
Use this for all your translation needs, natively, so outputs are always suited to everyone.

• New Teacher Starter Kit
• Best Tools for Teachers

Explaining AI use in schools to parents doesn’t have to be difficult, says Alana Winnick, Educational Technology Director at Pocantico Hills Central School District in New York and author of The Generative Age: Artificial Intelligence and the Future of Education.

Winnick, a Tech & Learning Innovative Award Winner, shares some guiding questions and advice for approaching AI in the classroom that have worked in her experience, though she stresses that every educational setting is different and teachers and school leaders tend to already know what’s best in their classrooms and schools.

Decide Whether AI Use Needs To Be Addressed With Parents?

The first step in discussing AI use in schools with parents is deciding whether or not that discussion even needs to happen, Winnick says.

Winnick asks rhetorically if teachers generally tell parents about every time they use a different technology tool. Since the answer is no, her follow-up question is: “Would you tell parents that you were using a calculator? So then would you have to disclose that you're using AI?”

In some settings, educators may want to explain AI tools being used in the classroom to parents, but it’s not always necessary, especially since a school with good vetting procedures for tech tools will already be taking student safety and privacy into account when approving any tech tool.

Educate Students and Their Parents By Teaching AI Literacy

Winnick has not heard of any parents worried about AI use in her district, but if such concerns do come up, she says the key to answering questions that might arise is to thoroughly educate students about their interactions with AI.

“It's about educating the students on AI literacy, so that way, if the parent asks them, 'Why are you doing that?' or 'Why are you using that?' the child can very confidently and eloquently say, ‘Oh, this is a tool, I use to help me with my writing; let me show you. I put my writing and I get feedback,' and the child understands it because we educated them," she says. "Then hopefully, that child will alleviate any of those fears or concerns with their parents. So the answer to most questions all comes back to education and educating the students and the teachers, and the parents.”

Always Answer Questions That Arise

Sometimes educating a student on AI literacy isn’t enough, and parents have questions about the technology and how its being used with their children.

“If a parent does come to a teacher or to the principal or the district and expresses concern, as with anything else, whether it's technology or something else, it's always important to address parents concerns,” Winnick says.

In this case, she notes that one strategy might be to remind parents that they and their children are likely interacting with a form of AI already. “At home, a lot of kids have Siri or Alexa right, and they’ll ask Siri or Alexa questions, so what's the difference if they're in class?” Winnick says.

Remind Parents of The Power of AI As A Tutor

Of course, another way you can explain AI use at schools to parents is by explaining the potential power of the technology as a tutor.

“In a public school, we are unable to provide every child with a one-on-one tutor-- it’s not possible,” Winnick says. “But with AI at least they can get all their questions answered. The AI can provide more challenging questions, or if they don't understand, it can break it down, and start providing more support than any other teacher would ever be able to do on their own. And it's also taking the data from that chatbot and providing it to the teacher, so the teacher knows who was struggling, or not struggling.”

AI can also personalize answers it uses to connect with student interests. For instance, if a student likes baseball, the AI might be able to frame an explanation using an analogy to that sport, Winnick says.

Share AI Success

For Winnick, AI’s efficacy in the classroom isn’t hypothetical; she’s witnessed it help many students.

​​”It's transformational for my English as a new language students,” she says.

Winnick adds that while AI tools that work with middle and high school students are often the focus, she says AI can work wonders with younger students as well.

“As a former early childhood educator, I spent my last year in a kindergarten classroom; those kids have so many questions, and a teacher does not have infinite patience to deal with the thousand questions they are getting all the time, but AI does," she says. "So when that child is super curious, you don’t want to burst that creativity; you really want to enable that. With AI, they can ask as many questions as they want.”

• 10 Free AI Trainings and Certifications for Educators
• GPT-4.5: What Educators Need To Know About The New ChatGPT Model

After exploring a number of education-related exercises with the free version of Gemini, I think that gap has closed. And if anything, in my usage, Gemini seems faster and has a smoother interface than OpenAI's ChatGPT.

Gemini also powers a number of education-specific AI tools for Google, including LearnLM, which recently left me very impressed, and a suite of Gemini for Education tools available through Google Education. For this review, however, I focused exclusively on the out-of-the-box version of Gemini 2.0 Flash.

Here’s how Gemini performed on a variety of tasks related to my work as a college professor and education in general.

Gemini and Class Prep

Class prep is a common use case for AI in education, so I started by asking Gemini to create an outline for a freshman composition class. I wrote: I am a college teacher. Teaching an introduction to college composition course for freshmen. I want to teach students how to write an argumentative essay with a strong thesis that makes a stance. I also want them to learn the basics of citation. Can you create a lesson plan, prompt for a 4-5 page argumentative writing assignment, and slideshow that I can discuss?

Gemini created a quick response that provided every element I asked for other than the slideshow. The prompt for the writing assignment was a good start and the bullet points it shared for a presentation were solid broadstroke outlines of what I often teach when I teach these classes. I don’t think this presentation is ready to go, but it could be a good starting point, especially for a new teacher looking to get started with how they might organize this type of class.

Gemini's Writing Advice

To see how a student might interact with Gemini, I asked it to help me eliminate the opinion elements from a piece I had written that was full of opinion statements and therefore not a good traditional news story.

Gemini offered helpful tips on the difference between opinion and news writing, and some solid rewrites of my opinion writing into a more neutral news story. In a perfect world, when working with students, Gemini wouldn’t rewrite the work but would suggest ways to rewrite it, yet on the whole, I think Gemini could help a student struggling to grasp news writing get back on track.

However, like all AI writing tools I’ve tested so far, I’m not impressed with its broader rewriting attempts. For instance, I thought its suggested rewrite for the opening of this story was lifeless and not really appropriate for a magazine story. Here’s the comparison:

My original: Google Gemini is the tech giant’s flagship generative AI model. When I first wrote about it back in 2023, Google was struggling to catch up to its generative AI rival. At that point I felt Gemini was impressive but was falling behind ChatGPT in the world of chatbot supremacy overall and as a teaching tool in particular.

Gemini’s rewrite: Google Gemini, the company's advanced generative AI model, has undergone significant evolution since its initial release in 2023. While early assessments highlighted areas for development, particularly in educational applications, recent advancements demonstrate its growing potential as a powerful tool for learning and teaching.

Gemini's Tutoring Ability

I asked Gemini to help me understand news writing (a topic I teach) and the Born rule of quantum mechanics (a topic I’m vaguely familiar with but trying to better understand).

In both cases, the information Gemini provided was helpful, and I did not see any inaccuracies. However, criticism of AI tutors in general is that these can act as answer machines for students and don’t engage in critical thinking to help them grasp topics -- and that criticism applies here.

Without elaborate prompting, Gemini acted as an on-demand search engine, quickly providing answers to various questions, which was still helpful and might be desirable in non-education settings but isn't perfect for teaching and deeper learning.

Some of the education-specific AI tools I’ve tried will engage in more of a dialogue, providing quizzes and questions that could force a student to put the concepts they are learning into action. The best example of this I have seen is actually the Gemini-powered LearnLM from Google, which I mentioned previously.

Gemini’s Writing Ability

One of the ways many educators interact with AI tools is when their students use any to submit a piece of writing.

To see what Gemini is capable of in this regard, I put one of my prompts into the tool and asked it to write a response. The response was good but didn’t sound like a student wrote it. AI is often lifeless, but Gemini sounded particular text-book and not student-like.

I also gave Gemini a creative writing prompt, which generated a result that was characteristically cliche-riddled corniness. For now, with the free version of Gemini, I’m confident I can still spot student AI-generated work and that my work as a writer isn’t in immediate peril.

Bottom Line

Gemini is a great, free AI tool that answers questions with speed, access to the internet, and an easy interface. Given Google’s prominence in the technology field, it’s a tool with which teachers will likely want to familiarize themselves. It serves as a nice alternative to some paid options out there, and has an easier-to-navigate website than some AI chatbot competitors.

Google and others offer more education-focused tools that are probably better in the classroom, but for educators just looking for a quick AI tool to use for activities such as class prep and brainstorming, Gemini is a great option.

• Google’s New AI Tutor LearnLM Is Trained On Learning Science and It Shows
• AI In Education: 4 Ways To Avoid Falling Behind The Curve

Here we speak with Travis True, Curriculum Technology Specialist for USD 501 Topeka Public Schools, about the benefits of implementing chatbots in the classroom, how teachers can set up chatbots, and why education should lean further into its innovative phase when it comes to technology.

Creating A Chatbot To Support Education

One of the main issues for teachers is the lack of time they have to dedicate to actual teaching. Most free time is often derailed by administrative busywork. However, a classroom chatbot can help to enhance the learning experience for students while also giving teachers an extra set of eyes, ears, and hands during the day.

“Chatbots, in the educational realm, give teachers an opportunity to create a personal learning environment for the student,” says True. “They get to manage what information is provided. They can upload their own information on what the chatbot is going to feed information from.”

And how can a chatbot work in education but outside of the classroom?

“At a school or district level it can allow building principals and district leaders to set up tools that parents or even staff can use to help find information quickly,” says True.

An example of this is True’s implementation of a high school handbook into a chatbot.

“One thing I’ve been testing and building is a chatbot that has one of our high school’s student handbooks uploaded to it,” says True. “The thought is that we put it on this website and make it available [for that high school]. Then it becomes a 24/7 tool that students and parents can use to ask questions about what’s in the handbook.”

Having the ability to easily access any specific information that you’re looking for can help save time while also allowing for better overall use of the resource.

Chatbots Inside the Classroom

Being able to pare down complex information is useful, but what about how chatbots can help students make strides with their education?

“Teachers can personalize those chatbots to kids who struggle with learning a concept,” says True. “They can also set up a chatbot for those kids who need a more advanced learning tool.”

Students who need a more specialized touch with their education can receive that attention with a chatbot. And, as True points out, creating one isn’t a difficult task for educators who have access to AI-enabled platforms such as SchoolAI or MagicSchool.

“The tools that the teachers have access to are pretty straightforward,” he says. “They enter instructions, standards, and how the chatbot is going to interact with the students. On the student side, there is a learning curve, because you have to ask good questions to get good information out of the chatbot. That’s a good skill for students, knowing how to ask good questions to get good answers.”

And in the event there needs to be a change in the information a chatbot provides (or what it responds to), it isn’t hard to keep up with changes.

“There’s always going to be some maintenance and some tweaking that needs to happen,” says True. “The tools that the teachers have access to (such as SchoolAI and MagicSchool) keep logs of every interaction. Teachers can go in and see how the students are interacting with those [tools].”

From there, a teacher can make the proper adjustments to have the chatbot serve students more effectively.

Embracing Game-Changing Edtech

The use of technology in education is not new, but adding AI has opened up so many doors to how we cater education for our students and teachers.

“This is a game-changer technology,” says True. “It’s moving a lot faster than Google did [when it came out]. We need to embrace it as educators and leaders. We need to embrace it because it is not going away. But we also need to teach kids how to use it correctly. It’s another pillar of digital citizenship. AI literacy, right up there with cyber bullying and using the internet correctly.”

Having all of this wonderful education technology is great, but it is important to teach students how to properly use what they have in order to give them the best chance of success going forward. Chatbots can do just that and much more.

• Do ChatGPT-Style AI ChatBots Help Students Learn? Yes, But There Are Caveats, Says Research
• AI Chatbot Friendships: Potential Harms and Benefits For Students

HOW WE TESTED: Three prompts were submitted to each platform over several visits, and the best and worst were selected for each gallery:1) Cat picking daisies; 2) Cat and mouse hanging out together as friends; 3) Cat chasing mouse across a field, with cat on the left side of the image and mouse on the right side. Both cat and mouse are running at top speed.

When I first surveyed free AI image generators in 2023, the technology was in its infancy. The handful of decent-looking AI-generated images were swamped by a sea of bizarre, odd, unintentionally humorous, and downright creepy efforts. Things have changed.

As everyone knows, the creative power of artificial intelligence has grown by leaps and bounds—and today's AI image generators reflect that overall improvement.

While some of the free AI generation tools I reported on in 2023 did not make the cut for this article, most are performing at a substantially higher level. And a couple of newer tools show great promise.

BEST FREE AI IMAGE GENERATORS

Bing Image Creator

A stripped-down AI image generator that nevertheless is capable of producing some decent images, Bing AI image generator automatically detects if you’re signed into a Microsoft account, making it easy to start creating images right away. Start with 100 credits, then apply rewards points accumulated within your MSN account for more.

Because it doesn’t offer a style menu, users must specify a style within the prompt, e.g.; “create image of a cat picking daisies in the style of a 35 mm camera.” Save, download, or customize your images. The platform applies controls to prevent potentially harmful violent or sexual image generation.

RESULTS: Bing frequently combined all of the prompt elements into a single nonsensical but rather amusing image. "Create image of a cat picking daisies in the style of a 35 mm camera" became a cat emerging from a 35mm camera with daisies in the background and a human operator in the foreground, delicately plucking a daisy from lens. Tweaking the prompts may help with this.

Deep AI

An easy-to-use site with a clean interface offering more than 100 styles of AI image generation, Deep AI provides an unspecified but generous number of free credits. AI model options consist of “standard,” “HD,” and “genius,” with genius being reserved for premium accounts. Five aspect ratios and two speeds are offered. Free users' images are saved in their account.

RESULTS: While DeepAI's images were generally good quality, it had a hard time interpreting the prompt. It often generated images of two cats instead of a cat an a mouse. In other cases, it took the phrase "hanging out" literally. It's curious that the model didn't recognize this common colloquialism. A carefully re-worded prompt may mitigate this tendency.

Freepik

Freepik offers an easy sign up with your Google or Apple account. Once logged in, free account users can select from three modes of image generation, 10 image aspect ratios, dozens of styles, plus color, camera, and lighting effects. Free users can generate and download 20 AI images daily, while premium features and the number of free credits remaining are clearly displayed. Your masterpieces can be found in the My Creations section.

RESULTS: Freepik's static renderings were generally fairly good, but it had more difficulty with the action scenes. Like some of the other platforms, it often depicted two cats instead of a cat and a mouse. It also didn't seem to understand the meaning of "chase," depicting animals running side by side in stead of one behind the other. Finally, its rendering of "top speed" was especially comical and inept.

Google Image FX

Image FX is one of several cutting-edge Google Lab tools that allow users to create and edit images, videos, and music with the help of AI. It’s easy to sign in with your Google account and start experimenting right away.

The clean uncluttered interface invites users to enter a prompt and select (if desired) one of the many styles. In contrast to most other tools, Image FX doesn’t display every possible image style in one screen. Rather, a handful at a time are revealed, with the option to click the "more" button if desired. Not sure which to choose? Try “I’m feeling lucky.” Additional options include aspect ratio and “seed,” which can be open or locked. A locked seed promotes more uniform results while the unlocked seed enables more variety. Users will quickly find their creations in the My Library link.

Currently, there are no paid subscriptions for Image FX, which means it’s blissfully free of urgent notices to upgrade your account. Nonetheless, there are limits. The Image FX FAQs note that each user is limited to a certain number of daily generations, in order to assure wide access. In my testing, I was able to generate more then 50 images using 18 different prompts without reaching my limit. I suspect the limit is based on demand and therefore flexible. Ideal for students aged 18 and over.

RESULTS: Image FX did an excellent job with the "cat picking daisies" prompt. In fact, it was the only platform that illustrated the action in an anatomically correct way. On the other hand, its depictions of cats chasing mice looked as if the cat and mouse were separately copied and pasted onto the background. But overall, Image FX was a very strong performer and well worth a try.

Imagine Art

ImagineArt’s free account provides 50 tokens every 12 hours, allowing about ten image generations. Users can choose from 11 aspect ratios, 14 styles, plus camera and lighting options while generating 1-4 images for each prompt. One particularly useful option is “text to png,” which generates your image as a .png file with a transparent background.

RESULTS: ImagineArt was at times confused between the cat and the mouse, sometimes depicting two cats or, amusingly, a cat and a rather cat-like mouse. Additionally, the word "mouse" apparently caused it to hallucinate Disney's Cinderella castle in the background of the chase scenes. Further, it solved the problem of "cat picking daisies" by simply adding a fifth limb. Overall, ImagineArt's free tools had a hard time producing accurate, well-rendered images.

Night Cafe

Night Cafe is like that goofy friend who's always doing something annoying but you love them anyway.

With its overly busy interface and frequent banners urging users to "unlock free credits," Night Cafe nearly didn't make it onto my list. Still, its unbridled enthusiasm and gamified interface is oddly appealing.

Free users have a choice from among several AI models and even the option to train a particular AI model to produce images in a preferred style. Granted, free users will only be able to generate five images from the custom model, but it's still an intriguing idea. Other options include five aspect ratios, choice of 1-16 images per prompt, speed of generation and randoms seeding.

Free credits can be earned by simply claiming any daily (five), creating a popular image, creating images two days in a row, and getting likes and comments on your images. Without even trying I accumulated 40 credits in a couple of weeks of occasional use. Daily challenges provide frequent opportunities to flex your creativity while the robust user community gives support, ideas and encouragement.

RESULTS: Night Cafe's produced several strong images of cats and mice hanging out as friends, as well as a few quite good renderings of the chase prompt. However, it often couldn't resist adding human-like anatomy to the cats picking daisies. Overall, it's worth a try, for the fun and the images.

• AI Art Generators in The Classroom: What You Need To Know
• 6 Ways Teachers Can Tell Students Are Using AI
• Best Free Image Editing Sites and Software

I asked GPT-4.5 to try its hand at creative writing, help with lesson planning, respond to prompts I use with students, and help me understand some concepts in physics with which I’ve long struggled.

After going through these tests, I'm not particularly impressed. To be sure, it’s a technological wonder that would have been gasp-inducing a few years ago. However, today OpenAI is something of a victim of its own success, and the latest and supposedly greatest GPT model is arguably outperformed by OpenAI and its competitors.

Here’s everything educators need to know about GPT-4.5 and how it might influence education.

What is GPT-4.5?

GPT-4.5 is the most advanced version of OpenAI’s flagship large language model. The original ChatGPT was powered by GPT-3.5. When GPT-4 was launched in the spring of 2023, I noticed a significant improvement over GPT-3.5.

This time around, things are more complicated. More players are now active in the AI space, such as Google, Claude, and Perplexity, and these are also regularly releasing updated AI models with interesting capabilities. The core GPT model also has competition from OpenAI’s own reasoning models, such as GPT-4o, which many say outperforms GPT-4.5.

How Does GPT-4.5 Do With Lesson Plans

I asked GPT-4.5 to create a lesson plan for my class. I told it I was a college professor teaching a freshman newswriting class and that I wanted students to learn the fundamentals of AP style reporting, objective reporting, journalism ethics, the inverted pyramid, and the difference between traditional journalism and many forms of new journalism, for instance, blogging.

GPT-4.5 gave a decent and broad lesson plan touching upon these topics. However, it wasn’t anything that I didn’t think previous GPT models or other AIs could have generated.

When I asked GPT-4.5 to suggest readings to go along with this lesson, it got really slow and seemed to crash. I had to exit and open a new browser and then ask again to get some suggested readings, but generating those readings was still slow. It would have been quicker and more productive to Google potential readings for a class in news writing.

How Is GPT-4.5's Writing Advice?

I put the first sentence of this story into the tool, and it suggested a rewrite that I thought subtly changed the original meaning. Overall, it wasn't a great edit:

Original:
“OpenAI recently released GPT-4.5 to ChatGPT Plus subscribers, and I began exploring the tool for its applications to education in general, and the college writing classes I teach specifically.”

Suggested revision:
“OpenAI recently released GPT-4.5 to ChatGPT Plus subscribers. Curious, I began exploring its potential applications in education broadly, and specifically within my college writing classes.”

How Is GPT-4.5 At Writing?

Many students continue to use AI for writing assignments in all levels of education. So it’s always a good idea to stay on top of what the latest AI models are capable of creating.

To assess this, I put in a prompt that I use for a freshman writing class I teach. The writing was technically solid but still seemed AI-generated. However, this AI writing is not as obvious as it was even a year ago.

I also gave it a detailed creative writing prompt and, as with other models, I was impressed and probably a little disturbed by the quality. Even so, it was definitely not good writing and still feels lifeless, but GPT-4.5 and other recently released models are cutting down on cliché use and making fewer weird mistakes. That said, it’s still strangely lifeless in both its tone and emotional impact.

GPT-4.5: Bottom Line

GPT-4.5 is an impressive piece of technology, but it doesn’t represent any major jump from other AI tools available both by OpenAI and other tech companies. A ChatGPT Plus subscription would not be worth it for this model alone.

Educators who are AI enthusiasts may want to play with this as they likely will find some specific uses that I didn’t. But for the AI-savvy, but obsessed, teacher, I don’t think there’s anything here they need to worry about missing. It also won’t really change the game in terms of how AI is being used by our students.

Ultimately, GPT-4 reminds me of the later iPhone releases. Sure, it got a little lighter, and improvements were made to the camera, but at the end of the day, your smartphone remained functionally the same.

• Google’s New AI Tutor LearnLM Is Trained On Learning Science and It Shows
• 5 Ways To Avoid AI 'Snake Oil,' According To ISTE Experts

“Whether you’re in person or you’re digital, there are icebreakers that you can use with technology and artificial intelligence,” says Kelli Erwin, Senior Instructional Design Manager at Learning.com. “A lot of people are still fearful of [AI]. So how can I take something as simple as an icebreaker, use AI and help break that barrier? We look at why we use icebreakers, how to do them, and what the benefits are.”

Combining AI and icebreakers can help improve engagement in the classroom, save time for teachers, and break down barriers altogether.

Using AI Tools in Icebreakers

If you’ve ever had to initiate a conversation with a group of people and did not know what to say, then you’re not alone. Icebreakers can help, but integrating AI with one can create an even more engaging environment to get students more involved.

So what kinds of AI platforms can be used to help icebreakers along?

“We actually use different tools for icebreakers,” says Erwin. “Things like Padlet, for example. That’s one of the tools that we use. One of the reasons I like Padlet is because it has a lot of neat tools built in, but it allows people to have a way to digitally communicate, even if they’re in person. We use a lot of the Adobe products, too. We use Adobe Express to get people engaged.”

These are two examples of technologies that can be used to introduce a different aspect to the traditional icebreaker concept.

How to Implement An AI Icebreaker in Your Classroom

Typical icebreakers have always been used in education, but adding AI tools to the process can make icebreakers feel brand new..

“Let’s say a teacher is introducing a new topic to students,” says Erwin. “Maybe the icebreaker is geared more toward the new topic. It’s helping take the fear out of a new topic. In Padlet, there’s a tool called ‘I Can’t Draw.’ It’s an AI tool and it makes you write a descriptive prompt so that the AI tool can go out and draw this image.”

Students can use this tool to better communicate what they know about a particular topic, but in a way that their chosen method of communication might succeed where simple conversation might fail.

“Students have to be very purposeful with how they write the prompt so that the image that is generated reflects what they’re intending,” says Erwin “Teachers can certainly use it as an assessment tool as well. Icebreakers can really be used anytime you need students to either show what they know, to collaborate, or break down any kind of barriers. It’s a fun way to open the door to discussions.”

Erwin offers another example of using an AI platform for an icebreaker to boost critical thinking skills.

“One [exercise] I do is called ‘It’s a Mystery to Me,’” says Erwin. “For this one, I usually use Adobe Express. I’ll pick a character and write a script for them, and I’ll read the script. The students have to listen to whatever the character says. It’s critical thinking but it’s also listening skills. So if teachers were to use something like that, it forces the students to practice their listening skills, but then have a conversation and think critically about what they think the character was talking about.”

By combining the ingenuity of AI tools with the social aspect of icebreakers, teachers can help students discover more about their work, their learning, and each other while also engaging them in ways that simple conversations cannot.

• Best Icebreakers for Back to School
• AI Starter Kit for Teachers

The good news is that there are many options for growing our awareness, knowledge, dispositions, and skills related to AI. This ranges from full graduate degrees in AI to professional development workshops. In this article, we focus on ten no cost AI training options that may be relevant to educators. Many of the options are connected to well-known and -used edtech tools, providing practical training when using.

Note: This list is in alphabetical order instead of rating these against one another.

1. Adobe

Generative AI for Creativity and Innovation offers an introductory generative AI course that’s been developed for using Adobe projects to innovate classrooms. The main topics covered include AI terminology, the power of prompt engineering, and ideas for using Adobe's AI-powered tools for creative classroom experiences. Within the course, real-world cases are presented for generative AI, which may help contextualize the content.

2. AI for Education

In approximately two hours, the Essential Guide to AI for Educators course provides training on using ChatGPT. The major topics that are covered include prompt engineering, ethics, and using ChatGPT in practice. A certificate of completion is provided at the end of the training.

3. Canva

Canva offers a 60-minute self-paced AI in the Classroom course for teachers. In it, content is presented with the aim of building confidence in Canva's AI-powered tools within teaching practices. This Canva AI training offers hands-on activities. Upon completion, a professional certification through Canva is provided. You can try integrating Canva's AI-powered tools with this Canva Lesson Plan.

4. Code Academy

The Intro to Generative AI 60-minute course provides introductory content on generative AI, so there are no prerequisite skills required. Three main areas that are focused on in the course include understanding what generative AI is, the different types, and the ethics of using it.

5. Code.org

Presented by Code.org, ETS, ISTE, and Khan Academy, AI 101 for Teachers offers a comprehensive online learning that focuses on teaching with and about AI. services. Companion Guides for Educators are provided to accompany the video training, which makes it more seamless to use the content learned in classrooms.

6. Common Sense Education

The Training Course: AI Basics for K-12 Teachers is a free foundational course focused on understanding the basics of generative AI while developing AI literacy. It also explores the impacts generative AI has had on education. As part of the course, reflecting on the connections between AI and teaching practices will be established.

7. Coursera

The four-module AI for Everyone is a six-hour beginner's level course. Modules include: "What is AI?" "Building AI Projects," "Building AI in Your Company," and "AI in Society." A shareable certificate that can be added to your LinkedIn profile, resume, or social media platforms is provided upon completion of the course.

8. Curipod

The AI-certified Curipod Educator goes beyond offering an introductory overview of AI by providing an opportunity for educators to become an AI-certified Curipod educator with a digital badge upon completion. Cuirpod specifically lists teachers, instructional technologists, curriculum content creators, and PD coaches as those for whom the training was designed. Using this Curipod Lesson Plan may be helpful in making connections to the Curipod AI-powered tools.

9. Google

Google has created Generative AI for Educators, which focuses on preparing educators to use generative AI for conversational prompting with tools such as Gemini and ChatGPT. The training should take approximately two hours, with exercises geared both toward writing prompts using AI tools as well as evaluating outputs.

10. Microsoft

The AI for Educators is a four-hour, four-module training offered by Microsoft for educators at the K-12 and higher education levels. In addition to providing content on the history of AI, the training includes preparing educators to use AI-powered tools such as Microsoft Copilot, Microsoft Edge browser, Minecraft, PowerPoint, and others. The training is aligned with the ISTE Standards for Educators and the UNESCO Standards for Educators.

The opportunities listed here are just a glimpse of options for free AI training, and certainly not exhaustive. Recently, one of our Tech & Learning contributors shared their experience with the Open AI and Common Sense Education's AI training as well as more detailed overviews of AI Literacy courses from Google and Adobe.

Try out one or more of the free AI training listed above. These opportunities to build professional learning can improve our approach in using AI for our instruction and assessment practices.

• 6 Top Free Edtech Credentials
• AI Starter Kit for Teachers

Here we speak with Eric Lawson, Director of Technology and Innovation at York School Department, about a unique method of introducing students to AI while also providing subtle guidance in its use and information about best safety practices.

What is AI “Speed Dating”?

The need for AI-focused learning is apparent, but there still is no overall agreed upon method of how to best teach children on its use. As AI continues to evolve, so too does the need to update what we know and how we teach.

While it might seem like a daunting task to keep up with an everchanging goal post, a solution may have been found: An AI lab specifically designed for student exploration and experimentation.

“It kind of grew organically,” says Lawson of the approach to introduce AI in bits and pieces. “We had a bunch of sixth grade teachers say that [they know kids are using AI and they need to get out in front of it]. How are we going to do this? About three years ago, they asked what we could do to explore different AI tools and to interview people in our community who were using AI for good work. That’s how we started.”

Although teachers were showcasing AI, Lawson says they wanted to create a situation in which students could hands-on experiment with it. Thus, a sort of AI speed-dating lab was born.

“I dreamed up this idea of having an AI lab, where we had 10 different stations with different tools all over the place,” says Lawson. “I created generic accounts. We read the students a riot act (don’t enter any personal identifiable information). ‘We’re not going to be using your accounts, so that information being fed into these AI prompts isn’t going to be yours necessarily. It’s just an experiment.’”

With a strong understanding of safety practices as well as parental consent, the students were able to explore the AI tools in short intervals.

“We had this kind of tool that they would go through and speed date different AI tools around the room, constantly switching every 10 to 15 minutes,” says Lawson. “Then they would reflect at a later time how a tool could be used appropriately. How a tool could save time. How it could enhance learning. They would also reflect on if a tool had any hiccups or concerns. How could that AI tool be used inappropriately, or how it could be harmful to students. They went through that in two different rotations to see everything that they could.”

Alongside well-known tools such as Magic School and School AI, each station that the students had access to provided a different form of AI interaction. These prompts would introduce an aspect of AI, ranging from how AI can be detected in imagery to using AI to create animation, video, and sounds. Students would then be asked to note how they would use these AI tools and what they learned in doing so.

Concerns with AI Use

Introducing students to AI can have benefits, but making sure they stay safe while using it is important.

“We didn’t want to use any AI tool that hadn’t signed our statewide student data privacy agreement,” says Lawson. “We were also very cognizant about any terms of services for any tools. Because these kids were right in that 12 age range. That’s why we got parental consent on some of these tools. That was a big reason for creating generic accounts. There was no information being passed through.”

What about the oversight of AI use for the students?

“Before we did the AI labs with the students, we had the staff run through it so that they could see what the tools look like, see if there were any hiccups,” says Lawson. “We even forced them to do an inappropriate prompt to see what the output might be. Then we had four adults in the room during the lab that were just walking through so that if [there was a technical issue or an output issue], we could come in to make sure that we redirected students to the appropriate use.”

How to Implement AI Speed Dating for Your Students

Keeping up with AI can be an arduous task, but approaches such as an AI lab can make it easier and fun to do. Here’s a few best things to keep Lawson says to keep in mind if you want to create something similar in your school.

“You need guardrails ahead of time,” says Lawson. “If you do not have an AI policy, there at least needs to be something spelled out. Parental consent was necessary for us, and the parents weren’t going to sign off until they saw where the guardrails were.”

Protecting student information is paramount.

“Ensure that you have enough generic accounts signed into all the tools ahead of time so that students aren’t accidentally going to sign in with their accounts and make it unsafe,” says Lawson. “Having that figured out ahead of time [is important].”

Knowing what the students wanted to learn about is key as well.

“We did it with a survey because we wanted to find out students' needs and interests,” says Lawson. “We then curated different tools based on those themes. The staff then went through them to understand which tools were going to be legit.”

By asking students about their AI related interests, vetting and learning the AI tools themselves, and creating a safe and secure learning environment, educators were able to introduce learners to a broad spectrum of AI-related tools able to help prepare them for the future.

• AI Starter Kit for Teachers
• AI Resources for District Leaders

Class Companion is designed for K-12 students and I teach college, but from my personal experience, it is an impressive AI tutor and grading assistant.

Here’s a closer look at what Class Companion is and what it does best.

What Is Class Companion?

Class Companion uses AI to provide students with personalized feedback in a few key ways. First, it provides personalized assessments of short-answer and essay questions. Second, it has a tutor named Ditto that can help students. Finally, it can also generate rubrics and assessment questions for teachers in addition to other features.

Class Companion utilizes OpenAI’s GPT-4 model to power its answers and tutoring features. However, the interface is much more user-friendly for teaching than using GPT-4 alone would be.

Additionally, instructors have control over every aspect of the process, from the types of answers you want the AI to look for and encourage to the number of opportunities students have to retry a response.

To start, teachers load a text into Class Companion -- for instance, a short story on which they want to assess students. Then, with or without the help of AI, they can create questions related to that story, and a rubric for the AI to assess students. Teachers can also, of course, override any AI feedback with which they don't agree.

What Are Class Companion's Best Features?

Class Companion's personalized feedback is what really stands out to me. Using AI as an instructor can feel like venturing into uncharted territory, but Class Companion allows the teacher to put in many guardrails and use the tool as they see fit.

Beyond this, I love how Class Companion supports mastery education philosophies. By default, the student is given multiple chances to submit an answer and can make changes to improve in real time based on feedback AI provides.

For instance, while testing the tool, I tried a sample question provided by another instructor about The Tale of Two Cities. The question: Describe the historical context provided by Dickens in the opening chapter. How does he portray the era in which the story is set?

I remembered the opening line of this novel and asked the AI tutor if it was okay to quote that in my response; it said it was. This is a great illustration of how an AI tutor could help a student who is unsure of how to proceed. Because it’s been a while since I read this book, I couldn’t recall many details from the novel beyond that famous 'best of times, worst of times' quote. So my answer wasn't the best.

The AI response celebrated that I’d included a quote but correctly encouraged me to add more details and specifics that addressed the substance of the question. If I'd been really taking this test, I would have gone back to the source material with some specific directions in mind but without the AI having provided the answer.

Setting up my own questions based on texts I loaded into Class Companion was also easy and quicker than I anticipated—though I would want to update the prompts I provided, based on student feedback.

How Much Does Class Companion Cost?

Class Companion is free to educators and their students, but many features, including LMS integration, are only available with a school or district subscription plan.

Class Companion does not publicly state the cost of these subscriptions, but school leaders can contact Class Companion to learn more and get a quote.

Class Companion Bottom Line: Would I Use It With Students?

The short answer is yes, provided it was approved by my institution, I would use Class Companion. It seems to be a helpful AI tool that keeps the focus on the educator and student agency while harnessing AI’s power to provide personalized feedback.

The longer answer is also yes, with the caveat that, as with all AI tools, I’d like to see more randomized control trials in which groups of students take the same class with and without an AI tool such as this. I think it’s extremely unlikely that a tool such as Class Companion would hinder a student’s educational progress the way AI tutors deployed in less pedagogically sound ways have in the past, but I’d like to quantify how much it helps.

By understanding the big-picture impact of AI tools such as Class Companion, teachers and school leaders can better assess which AI tools are worth the investment.

• AI In Education: 4 Ways To Avoid Falling Behind The Curve
• How YOU Can Avoid Investing In Bad AI and Risking Student Data