Insights from 50 AI-in-Edu Experts
What 11,844 pages of commentary reveal about AI in Education
I read a lot about AI. To date, I have 1,126 bookmarks related to AI alone. But that pales compared to the 6,478 AI-related emails in my Gmail account.
I love to read, but clearly, I can’t read everything. So, over time, I’ve developed some strategies to help me organize, summarize, analyze, and extract information from my emails. With varying success, I’ve experimented with Gemini, Zapier, Google Takeout, Python, and ChatGPT in that effort.
This week, I used Google Takeout to export all 1,425 emails labeled “AI Edu Commentaries” and uploaded them to ChatGPT. These are commentaries from a range of impressive AI in education practitioners, academics, researchers, consultants, and tech leaders. Mostly newsletters and blogs, these 1,425 emails constitute 11,759 pages of text, or roughly three million words. For good measure, I threw in 85 pages of YouTube transcripts from other notable AI-in-education experts.
Why? I wanted to glean insights from this abundance of expert commentary.
I wanted answers to these and other AI-in-education questions:
Are there any frameworks or models repeatedly referenced for integrating AI into instruction?
Which classroom applications of AI are described as the most beneficial or impactful?
How do the authors/speakers suggest using AI to support differentiated or personalized learning?
How is AI being used to support student creativity, critical thinking, or collaboration?
50 AI-in-Education Experts
Most of the commentaries come from education professionals who focus primarily—or significantly—on AI in their work. That said, I added a few experts who address AI as part of a broader push for educational reform or a particular academic focus. I also included prominent “tech voices,” like Google’s and Microsoft’s Head of Global Education. I strove to create a cross-sample of AI-in-education perspectives that might elicit more insightful information.
Here are the 50 experts I selected
While virtually all of these individuals advocate using AI in education, their priorities sometimes differ and they hold varied viewpoints and perspectives on implementing AI in teaching and learning.
With that in mind, let’s start with the questions and responses:
In partnership with Essay Grader
Essay Grader is a straightforward, powerful tool that uses AI to take the load off your shoulders and will completely change the way you approach grading!
Every feature puts teachers first, with a focus on simplifying repetitive tasks like grading to improve productivity - all the while letting teachers be fully in control.
Streamline grading, provide detailed, actionable feedback to your students, and maintain complete control over your assessments - all in one intuitive interface.
Reduce grading time by 80%. Grade your entire class's essays in 2 minutes or less and deliver high-quality, specific feedback to your students.
Ensure fair essay grading by removing bias and aligning with state standards like Texas STAAR, Florida BEST and California CCSS.
Join Essay Grader and its ever-expanding network of 60,000+ educators, and you’ll connect with others who share your passion and commitment.
1. Are there any frameworks or models repeatedly referenced for integrating AI into instruction?
ChatGPT’s full scan of all 1,917 emails shows that “AI literacy” appears 538 times. All told, AI literacy is a major recurring theme across this entire body of educational commentary.
ChatGPT’s analysis of over nearly 3 million words from prominent AI-in-education commentators reveals that AI literacy is a frequently cited concept and a cornerstone of current thinking about AI integration in K–16 education.
AI literacy is most often discussed in the context of student skill development, particularly prompting, hybrid writing, and critical thinking. Other references involve teacher training, institutional planning, and calls for national policy support.
Emerging from these texts is a vision of AI integration shaped more by practical experimentation and professional intuition than by traditional frameworks. Educators are urged to equip both themselves and their students with the skills to engage AI thoughtfully, ethically, and creatively in daily academic work.
Key Sample Passages (Selected)
“We must better integrate intentional AI literacy into our curricula.”
“Teaching students to ‘code switch’ between AI-enhanced writing processes and traditional approaches ensures they maintain fundamental writing skills while also developing their AI literacy.”
“Introducing two AI literacy courses for educators… designed to build skills and foster community learning.”
“While they understand the relevance of generative AI, they underestimate the critical value of direct AI literacy and prompt engineering skills in future careers.”
“A draft executive order aimed at K–12 schools would make AI literacy a national priority…”
I doubt you’re surprised that AI literacy is a center of attention. Yet, I was surprised that traditional instructional models such as SAMR, TPACK, or UDL are virtually absent from the discourse. It’s a noteworthy absence and indicates that attention is focused squarely on new, emergent dynamics. So, rather than invoking traditional frameworks like TPACK, experts like Nick Potkalitsky and Stefan Bauschard call for emergent models rooted in iteration, feedback fluency, and scenario-based design. There is a markedly pronounced Process vs. Product emphasis at play: Dr. Justin Reich, Dan Meyer, and others stress reframing instruction to assess the learning process rather than just final outputs, encouraging AI as a partner in iteration and inquiry.
AI Literacy: Areas of Agreement and Disagreement
As a follow up, I asked ChatGPT if there much consensus over what 'AI literacy' means, how it should be defined, and how it should be cultivated.
Here is what it identified:
Points of agreement:
AI literacy includes both understanding and using AI tools effectively, safely, and ethically (Law, Hardman, Spencer, Reich).
It must integrate:
Tool fluency: Knowing how to prompt, evaluate outputs (Hardman, Potkalitsky)
Ethical awareness: Privacy, bias, misinformation (Spencer, Reich, Bauschard)
Process transparency: When and how AI was used (Hooker, Mollick)
Points of disagreement:
Is it skills-based or mindset-based?
Some advocate for flexible, evolving mindsets — curiosity, creativity, critical questioning. (Hardman and Law)
While others emphasize concrete skills and explicit instruction — how to prompt, verify, and revise. (Spencer and Warner)
Should it be treated like digital literacy or something broader?
Some say AI literacy deserves its own category, because it reshapes cognition and creativity. (e.g., Darren Coxon, Richards)
Others frame it as an extension of digital/media literacy (Warner, Cuban).
Some like Mollick, and Rostan advocate a hybrid approach, where competencies evolve through adaptive mindsets and real-world practice. I fall into this camp. AI literacy is both skills-based and mindset-based. It’s not just about knowing how to prompt. It’s about knowing when and why to use AI — and when not to.
2. Most Impactful Classroom Applications of AI (Based on Frequency and Context)
According to ChatGPT’s analysis, the most frequently referenced and evidently impactful uses of AI in classrooms include:
Reading and Comprehension
AI tools are used to simplify or scaffold texts for different reading levels, summarize complex readings, and generate comprehension questions.
Project-Based Learning
Educators frequently mention AI as a partner in interdisciplinary projects, helping students research, simulate roles, or create multimedia outputs.
Personalized Instruction
AI is referenced often in relation to “differentiation,” “scaffolding,” and adaptive “feedback” — signaling strong interest in AI for meeting diverse learning needs.
Assessment and Feedback
Many authors and speakers describe AI’s usefulness in providing formative feedback, automating low-stakes quizzes or reflections, and helping teachers manage grading load.
Lesson and Instructional Planning
AI tools are regularly used by teachers to generate lesson plans, adapt content for standards, or tailor activities to student interests.
Student Writing Support
AI is heavily used to assist students in brainstorming, drafting, revising, editing, and summarizing written work.
—I asked ChatGPT a follow-up question—
What AI tools are recommended for the implementation/achievement/development of any of these objectives, and why?
Here is how it responded:
This table notwithstanding, most of the commentaries in the 1,917 emails and ~85 pages of YouTube transcripts are general or theoretical and devoid of specifics on AI tools. Given the data, it’s striking that only ~4% of the total discourse includes practical guidance about using AI tools. Moreover, if you remove discussion of well-known AI chatbots, and a few popular AI teaching tools, just 0.6% of the entire corpus (~17,700 words) discusses AI tools.
This is striking considering how many experts on this list are practitioners or former practitioners. Tools like ChatGPT, Claude, Gemini, MagicSchool, Diffit, Canva, and NotebookLM are mentioned with some frequency, but usually in the context of general benefits (e.g., saving time, brainstorming) rather than specific curriculum practices. For teachers begging for more practical guidance, that’s little solace.
3. How do the authors/speakers suggest using AI to support differentiated or personalized learning?
Across the dataset, there’s a clear pattern: AI isn’t just about delivering different content to different learners — it’s about reshaping the pathways through which students learn. Experts like Bauschard and Hardman emphasize iteration, feedback cycles, and AI as an “adaptive learning partner.” Others, like Watkins and Hudson, highlight AI’s emotional neutrality — which can make feedback safer and more actionable.
i. Customizing Content by Reading Level, Language, and Interest
Tools mentioned: Diffit, MagicSchool, SchoolAI, TeachAI, Eduaide
Who says it: Alicia Bankhofer and Philippa Hardman advocate tailoring content to individual learner profiles. Carl Hooker uses Diffit to adjust texts for reading levels and home languages.
Use case:
A 6th-grade teacher generates three versions of the same article — one simplified for ELL students, one standard, and one enriched — each in the student’s preferred language.
ii. Prompt-Based Personalization and Thought Partnering
Tools mentioned: ChatGPT, Claude, Gemini, Notion AI
Who says it: Ethan Mollick uses AI to co-create inquiry tasks. John Spencer frames AI as a “cognitive partner” helping students brainstorm and work through tough concepts.
Use case:
A student prompts ChatGPT to compare Māori and Navajo environmental perspectives, refines the output with feedback, and develops a cross-cultural analysis.
iii. Automating Differentiated Lesson Plans and Scaffolds
Tools mentioned: MagicSchool, Eduaide, TeachAI, Khanmigo
Who says it: Marc Watkins uses AI to generate tiered lesson variants. Jeanne Beatrix Law stresses chunked scaffolds for neurodiverse learners. Dan Fitzpatrick highlights philosophical tailoring using Eduaide.
Use case:
A biology teacher creates a 3-tier lab guide on osmosis — one version with vocabulary scaffolds, one with open-ended challenges, and one with extension tasks.
iv. Providing Individual Feedback and Check-ins
Tools mentioned: SchoolAI, ChatGPT with custom prompts, Khanmigo
Who says it: Eric Hudson values AI’s “emotion-free” feedback for hesitant learners. Stefan Bauschard promotes low-stakes AI feedback before peer review.
Use case:
A student uploads a draft to SchoolAI and receives grammar, structure, and tone suggestions — revising before sharing with peers or the teacher.
v. Letting Students Choose Paths or Products
Tools mentioned: ChatGPT, Adobe Firefly, Canva, Google Slides with AI plugins
Who says it: Michael Spencer encourages co-designing assessments. Bryan Alexander and John Warner support multimodal expression powered by AI.
Use case:
Students pick from making a podcast, writing an op-ed, or designing a visual explainer — all using AI tools to scaffold and extend their creative process.
Vander Ark’s model of personalized pathways aligns well with AI-powered student choice — whether it’s picking the medium of expression or designing their own learning trajectory.
—I asked ChatGPT a follow-up question—
What innovative teaching strategies involving AI are described by the email authors/YouTube speakers?
The dataset presents a “compelling set” of innovative teaching strategies that go far beyond "AI as time-saver" and instead reimagine pedagogy around co-creation, inquiry, reflection, and process over product.
Here is a curated synthesis of the “most innovative” AI-infused teaching strategies discussed by these thought leaders:
1. “Human vs. AI” Challenge Tasks
Carl Hooker
Students go head-to-head with AI in creative challenges, like brainstorming picnic items or writing personal stories, then reflect on what AI missed.
Why it’s innovative: Emphasizes human creativity and nuance; builds critical AI literacy through comparison.
Pedagogical impact: Encourages metacognition, critical thinking, and healthy skepticism about AI outputs.
2. Reverse Engineering AI Outputs
Michael Spencer, Bryan Alexander
Teachers give students AI-generated responses and ask them to critique, revise, or improve them using classroom knowledge.
Why it’s innovative: Encourages students to act as editors, fact-checkers, and revisers.
Pedagogical impact: Builds both AI literacy and domain content mastery.
3. Process-Driven Grading Models
Jeanne Beatrix Law, Dan Meyer
Assignments now require students to document how they used AI, why, and what they learned from it — making the learning process visible.
Why it’s innovative: Moves beyond “final product” grading to learning narratives.
Pedagogical impact: Deters academic dishonesty, rewards reflection, and humanizes assessment.
4. Equity-Driven Access to Scaffolds and Translation
Dr. Craig Hansen, Jeanne Law, Darren Coxon
AI generates scaffolds, translations, and visual aids for neurodiverse students or English learners — customized on the fly.
Why it’s innovative: Embeds accessibility into daily instruction.
Pedagogical impact: Advances inclusion and just-in-time personalization.
5. AI-Enhanced Performance Tasks
Dan Fitzpatrick, Michael Spencer
Students use AI to simulate historical arguments, design ethical decision trees, or create multimedia artifacts that require judgment, synthesis, and creativity.
Why it’s innovative: Moves past recall and regurgitation to higher-order thinking.
Pedagogical impact: Aligns with deeper learning goals and authentic assessments.
6.AI-Prompt Debugging + Prompt Jams
Nick Potkalitsky, Alicia Bankhofer
Students work in pairs to diagnose why a prompt “failed” and improve it iteratively — akin to “debugging code.”
Why it’s innovative: Teaches prompt engineering as a literacy.
Pedagogical impact: Fosters resilience, iteration, and a design mindset.
4. How is AI being used to support student creativity, critical thinking, or collaboration?
i. Creativity
ChatGPT reports that AI is most frequently used to spark and extend student creativity through:
Idea generation and brainstorming
Tools like ChatGPT, Claude, and Canva Magic Write help students generate story ideas, artistic prompts, or new angles for research projects.
Marc Watkins encourages students to use AI for rhetorical experimentation, seeing it as “a new form of creative play.”
Dr. Caitlin Tucker promotes AI writing tools for creative revision and stylistic experimentation.
Multimodal creative production
Students use AI image generators (e.g., DALL·E, Adobe Firefly) for storytelling and visual narratives.
Voice and language experimentation
Paul Matthaeus and others note students are using voice-cloning and language transformation tools (like HeyGen) to present stories in multiple voices and languages.
ii. Critical Thinking
AI is being framed not as a replacement for critical thinking, but as a catalyst for deeper student engagement with evidence, bias, and reasoning. Common strategies include:
Critiquing and revising AI output
Educators like Dr. Justin Reich, Dan Meyer, and Terry Underwood encourage students to challenge AI responses, identify hallucinations, and rewrite outputs with improved logic or evidence.
Dr. Jeanne Beatrix Law promotes AI-supported Socratic questioning: students prompt AI, then interrogate its responses.
Fact-checking and source triangulation
Eric Hudson and Nick Potkalitsky describe workflows where students compare AI-generated content to human-written texts or real-world data, especially in history or civics.
Meta-awareness of AI logic
Tools like NotebookLM and Perplexity AI are used for tracing reasoning, analyzing source use, and understanding “how AI got there” — a practice championed by Philippa Hardman and Ethan Mollick.
iii. Collaboration
AI is being used to enhance peer and student-teacher collaboration, particularly through:
AI as a "third collaborator" or rehearsal partner
Several educators (e.g. Caitlin Tucker, Carl Hooker, Michael Spencer) describe students using AI to rehearse arguments or prepare drafts before peer review, helping level the playing field.
Shared inquiry and co-editing
Dr. Craig Hansen and others report AI-supported peer co-editing: students bring AI-generated or AI-edited versions of their work into workshops and collaborate on refinement.
Collaborative design tasks
In makerspaces or project-based learning, students are using AI (e.g., MagicSchool AI, Diffit) to co-develop project plans, storyboards, and checklists, encouraging division of labor and negotiation.
AI-facilitated multilingual communication
In international or multilingual classrooms, tools like HeyGen, DeepL, and ChatGPT's translate function are used to help students collaborate across language barriers (as discussed by Dina Ghobashy).
Emerging Themes Across All Three Areas
AI as a co-creator (rather than an answer machine): This shift supports creativity and reflection.
AI as a feedback tool: Fast, emotion-free feedback helps encourage iteration and improvement.
AI as a scaffold for equity: Used thoughtfully, it helps students who may struggle with expression, confidence, or language barriers engage meaningfully in collaboration and creation.
Potkalitsky and Rostan, in particular, push us to think of AI not as a passive tool but as a dynamic partner — one that sharpens reflection, reveals assumptions, and scaffolds metacognition.
Conclusion
This exercise largely confirms what many of us are feeling: We’re in the midst of a deep rethinking of teaching and learning. Collectively, these 50 voices and 3 millions words represent a shifting center — from product to process, from teacher content delivery to student AI partnership. But most striking is that few prescribe tools. Few quote legacy frameworks. Instead, they wrestle with practice.
I feel for teachers who want more specific tool guidance. But this is a work toward a pedagogy of co-creation, iteration, and transparency. What we lack now in specificity, we are gaining in vision. So that’s the challenge in front of us — reimagining the classroom, with AI beside us, not ahead of us.
AI in Edu: News, Tools & Views is sent to all subscribers on Fridays.
Find more AI tools, views, and how-to’s at tomdaccord.com
Find my book AI Tools & Uses: A Practical Guide for Teachers at Amazon
*Learn more about Essay Grader