Last week, one of my students asked an AI chatbot why World War I started. The answer it gave wasn’t wrong — but it also wasn’t the point.
I’ve been using AI in my AP World History classes for almost a year now, and I’m still figuring it out. Some days it feels like a superpower. Other days, it raises more questions than it answers. That’s actually kind of perfect for a history class, in my opinion.
This is, again, my attempt to think out loud — pulling in voices from classrooms, research, and the broader education conversation — about what it actually looks like to teach with AI, not just teach about it.
The Big Picture: We’re Past the Tipping Point
Let’s start with the numbers, because they’re striking. According to a 2025 Microsoft report, 86% of education organizations now use generative AI — the highest adoption rate of any industry. Student use for school-related work jumped 26% in a single school year. Teacher use rose 21% over the same period.
This isn’t a wave that’s coming. It’s already here. The question isn’t whether AI belongs in classrooms — it’s what role it plays, who’s guiding that role, and whether it’s actually helping kids learn to think.
Key stats:
– 86% of education orgs now use generative AI
– 62% increase in test scores with AI-powered instruction
– 87% of principals worry AI hurts critical thinking
Sources: Microsoft On the Issues (2026); Engageli AI Statistics (2026); Education Week / Alyson Klein (2025)
What I’m Actually Doing in My Classroom
Here’s the honest version: I started small. I used AI to help me differentiate primary source documents — taking a dense excerpt from a Ming Dynasty trade record and generating three different reading-level versions so every student could actually engage with the same source. It saved me a couple hours of prep. More importantly, it meant no student was shut out of the historical conversation.
I started experimenting on my own, creating more and more complex activities and building better bots. I used ElevenLabs to create an AI tutor for my AP classes that had my voice. I used the textbook as the knowledge base, and the students were able to have natural conversations with me, their teacher, at any time.
I started having students use AI tools to generate draft historical arguments — a thesis for a DBQ, say — and then critique what the AI produced. What’s missing? What’s oversimplified? What would a real historian push back on? Suddenly the AI wasn’t doing the thinking for them. It was giving them something to think against.
That shift — from AI as answer machine to AI as sparring partner — changed everything.
“The real task is not to contain AI, but to design learning experiences that develop students’ historical thinking, ethical reasoning, and human judgment in a world with AI.”
— Disrupted History Blog, 2025
What’s Actually Useful for History Teachers
There are a lot of AI tools out there. Here are a few that I’ve either used or that the community is genuinely excited about — specifically for social studies contexts, as curated by Dr. Med Kharbach of Educators Technology (Feb 2026):
Humy.ai — Students can interview, debate, and role-play with AI-powered historical figures. Lesson plans included. The AI grades assignments and integrates with most LMS platforms. This one is built specifically for social studies — it shows.
Diffit — Paste in any primary source — a Silk Road merchant’s account, a Cold War speech — and Diffit automatically generates multiple reading levels, comprehension questions, and vocabulary lists. Essential for diverse classrooms.
VictoryXR — Over 150 immersive field trips: D-Day beaches, the Berlin Wall, Egyptian temples. An AI Tutor feature lets students ask questions in context. For a World History class, this is genuinely exciting.
And worth bookmarking: Tom Daccord’s curated list of AI tools for history teachers is a solid resource updated regularly by a practitioner who really gets the discipline.
Does AI Help Students Think — Or Think For Them?
This is the tension at the center of everything right now. Education Week’s Alyson Klein reported last October that 87% of principals worry AI reduces critical thinking. 100% cited cheating concerns. Those numbers are hard to ignore.
But here’s the interesting part: students don’t see it the same way. Only 45% of high schoolers worry about skill erosion from AI use. There’s a genuine perception gap — and it matters, because both sides are probably a little bit right.
“There’s a little bit of a perception gap with students thinking ‘this is grand!’ and teachers thinking ‘this is not really helping them.'”
— Jessica Howell, VP of Research, College Board (via Education Week, 2025)
Chad Sussex, an assistant principal in Iowa, offered what I think is the right frame: teaching students to critically evaluate AI outputs — rather than just accept them — could be the thing that actually closes that gap. In a history class, that means asking: What would a historian say about this? What’s the AI leaving out? Whose story is missing?
That’s not a limitation of AI. That’s a pedagogy.
Media Literacy Is Now an Ethical Emergency
I want to spend a moment on something that doesn’t get enough airtime in conversations about AI and K-12 education. Colleen Kenny, writing on LinkedIn, calls what we’re living through “content chaos” — a collapse of shared frameworks for evaluating information. Media trust in the U.S. now sits at 28%, according to Gallup. And the average person genuinely cannot tell the difference between real and AI-generated video.
For a history teacher, this hits close to home. We’ve always had to help students navigate propaganda, bias, and perspective. Now we’re doing that while the landscape of what constitutes “real” is actively shifting under our feet.
“Education must prioritize multilayered literacies — cognitive, emotional, and embodied — as foundational, not supplementary, to sustaining democracy.”
— Colleen Kenny, LinkedIn (2025)
What I appreciate about Kenny’s piece — beyond its substance — is that she disclosed her own AI use in writing it: Gemini for structure, Claude for style, Perplexity for fact-checking. That kind of transparency is what we should be modeling for students. The tool isn’t the problem. Using it without accountability is.
In my lifetime, we’ve gone from having curated, reliable sources in our libraries where doing research was about consuming information and using it to create new content. The internet made information available on mass scales, but it also introduced misinformation and disinformation at mass scales, as well. Learning how to think through issues and discern the good from the bad in terms of information is now much harder — and much more important than ever.
This Isn’t Just a Tech Question
The OECD’s Digital Education Outlook 2026 puts it plainly: GenAI can support learning when guided by clear teaching principles. But without that guidance — without intentional pedagogy behind the tool — you just get performance without learning. Students who can produce a polished-looking essay without understanding anything in it.
UNESCO’s ethics framework adds something important: AI systems trained on historical data often reflect and reproduce societal biases. For a history class, that’s not a bug to warn students about — it’s a lesson. What does it mean that an AI trained on predominantly Western sources might give a very different account of the Mongol Empire than a Chinese historian would? That’s the kind of question that makes history come alive.
The equity dimension matters too. Not every school has equal access to AI tools, reliable internet, or devices. As we get excited about what AI can do, we can’t let that excitement widen the gap between well-resourced and under-resourced classrooms.
What I’m Taking Into My Classroom This Week
- Use AI to generate a “flawed” primary source analysis — then have students find what’s missing, biased, or oversimplified.
- Ask students to disclose how they used AI on any written assignment. Model it myself in class.
- Try Diffit on one complex document this week. See if differentiated access changes the discussion.
- Bring in a real AI-generated video clip and a real clip on the same historical event. Ask: how do you know which is which? Does it matter?
- Keep asking the OECD’s question: are my students learning, or just performing?
- I’m also using Claude Code to create a Build Up to WWI App that can engage the kids by gamifying important information.
Thanks for reading. This newsletter is a work in progress — just like my AI practice. If something here sparked a thought, I’d genuinely love to hear it. The conversation is the whole point.
Sources referenced:
– Colleen Kenny — “Content Chaos: Why AI Makes Media Literacy Education’s Most Critical Skill” (LinkedIn)
– Alyson Klein — “Teachers Worry AI Will Impede Students’ Critical Thinking Skills” (Education Week, 2025)
– Dr. Med Kharbach — “AI Tools for History and Social Studies Teachers” (Educators Technology, Feb 2026)
– OECD Digital Education Outlook 2026
– Microsoft — “Teaching in the AI Age” (2026)
– Tom Daccord — AI Tools for History Teachers
– Disrupted History — “From Levels to Learning: Rethinking AI Assessment in History Departments” (2025)
