By David Jacobson | AP World History, Shanghai American School
I had a student last semester ask me a question that I couldn’t immediately answer. Not about the Industrial Revolution or the Silk Road — about whether the source she’d found using an AI research tool was real. She’d cross-referenced it, found a matching title, but the quotes were subtly off. The text existed. The quotes didn’t. She caught it. I hadn’t taught her that specifically. She’d just developed the instinct.
That’s the thing about teaching in an AI-saturated environment: the skills we’re building now are ones we couldn’t have predicted needing five years ago. And nowhere is that clearer than in international schools, where the stakes of getting this right feel particularly high.
China Is Moving Fast — and Deliberately
While much of the debate in the United States remains stuck at the level of “should students use AI at all?”, China has already settled that question. AI education is now state-mandated in primary and secondary schools, with a full curriculum rollout underway. This isn’t peripheral — it’s treated as national infrastructure.
What’s interesting, though, is that the conversation inside Chinese classrooms isn’t purely technological. At the Suzhou Experimental Primary School in December 2025, six teachers presented their approaches to AI integration at a school forum — and the questions they raised were deeply humanistic. Researcher Thomas Hatch, who documented the forum, highlighted a striking comment from Ms. Hu, a data teacher: “AI can give us a clear diagnostic map. But a prescription is our teacher’s duty.” Ms. Wang, who teaches art, put it even more memorably: “AI is a museum that helps us find inspiration, not a print shop.”
That last phrase stopped me when I read it, because it’s exactly right — and it echoes something I’ve been working through in my own AP World History classes. The problem isn’t that students are using AI. The problem is when they use it as a print shop instead of a thinking partner.
What This Looks Like in an International School History Classroom
At Shanghai American School, I’ve been experimenting with how AI fits into the work of historical analysis. What I’ve found is that AI is genuinely useful for certain things — generating timelines, summarizing background context, producing first-draft outlines — and genuinely problematic for others.
The moment I push students toward primary source analysis, the limitations show up fast. AI can tell a student what the Treaty of Nanjing said. It struggles to help them understand what it felt like to be a Qing official signing it, or how that moment reverberates into contemporary China-West tensions. That’s interpretive work. That’s the job.
So I’ve started using AI strategically as a scaffold, not a scaffold that holds the building up permanently but one that gets students into the material faster so we can spend class time doing the harder thinking. The AI handles what’s mechanical. We do what’s human.
Across international schools in China, AmCham China has documented similar evolutions — schools moving from early-stage experimentation to structured GenAI policies, professional development programs, and student-facing ethics frameworks. This isn’t reactive anymore. It’s intentional design.
Media Literacy: The Skill We Can’t Afford to Skip
Writing for LinkedIn, Colleen Kenny makes an argument that’s been sticking with me. She describes what she calls “content chaos” — the collapse of shared frameworks for evaluating information in an age of algorithmic curation and AI-generated content. As she puts it: “Education’s most urgent task in the AI age is equipping the next generation globally to navigate content chaos without losing their grip on a shared reality.”
Trust in media in the U.S. has fallen to 28% (Gallup). The situation in internationally mobile communities — where students consume content from multiple national contexts, in multiple languages, shaped by different algorithmic ecosystems — is even more complex. My students are simultaneously navigating Chinese social media, Western platforms (via VPN), and academic sources. The question of what’s real, what’s reliable, and what’s been subtly warped is one they face every day.
Kenny’s argument, and I think she’s right, is that media literacy isn’t a supplemental skill anymore. It’s the foundation. And AI tools, paradoxically, are both the cause of the problem and — if used well — one of the best tools for teaching students to navigate it.
The Pedagogy Question: What Does AI-Enhanced Teaching Actually Look Like?
Here’s where it gets interesting for teachers. A recent global survey found that 85% of teachers used AI in the 2024-2025 school year, and 69% said AI had improved their teaching methods. But only 30% feel confident using these tools. That gap matters.
The UNESCO King Hamad Prize for 2026 specifically recognizes projects demonstrating how AI can encourage critical thinking, imagination, and creativity — not replace them. That framing is important. It pushes back against the fear that AI flattens learning into automated content delivery.
For history teachers, the pedagogical spectrum looks something like this: at one end, AI can handle the lowest-order tasks — generating vocabulary lists, summarizing chapters, producing quiz drafts. At the other end, the hardest interpretive work — reading between the lines of a source, arguing for a historical claim, making sense of conflicting accounts — stays human. What AI can do is compress the time it takes to get students ready for that harder work, and give teachers more time to actually teach.
At the Suzhou forum, a kindergarten teacher named Ms. Sheng asked a question that cuts right to it: “Will AI strengthen or interfere with teacher-child interactions?” That’s not a kindergarten question. That’s the question for every level of education right now.
My answer, from where I’m standing in Shanghai: it depends entirely on how you use it. AI used to replace interaction produces worse learning. AI used to create more space for interaction — by handling the mechanical stuff so teachers can show up more fully — produces something genuinely better.
Ethical AI: Not a Side Conversation
International schools have an added layer of responsibility here. Students who graduate from schools like mine will move through multiple national contexts. The AI systems they use in China operate under different regulatory and surveillance frameworks than those they’ll encounter in the UK, Australia, or the U.S. Understanding how those systems work — who owns the data, what it’s used for, what the assumptions are baked into the model — is part of being an educated global citizen.
Recent research from Frontiers in Education identifies the key ethical concerns for educators: data privacy, algorithmic bias, and educational inequality. These aren’t abstract worries. In an international school context, where students come from dozens of countries and learning differences are real and varied, the question of whether AI tools systematically advantage some students and disadvantage others is one we have to take seriously.
What I tell my students: AI is a tool built by people with particular assumptions, trained on data that reflects particular histories and power structures. It’s not neutral. Analyzing an AI’s output with the same critical lens you’d apply to any historical source is exactly the right instinct — and it’s the same instinct good historians have always used.
What’s Working, What to Watch
A few things I’m tracking as this semester moves forward:
The OECD’s AILit framework is building toward a PISA 2029 assessment of AI and media literacy. That’s a signal: AI literacy is becoming a structured global standard, not an elective skill.
Adaptive learning platforms are being adopted by 71% of higher education institutions by 2026, up from 34% in 2023. K-12 schools, especially internationally, are following quickly.
And the teacher confidence gap remains the most urgent structural problem. 85% of teachers using AI, 30% feeling confident. That’s not a technology problem — it’s a professional development and institutional support problem. Schools that invest in training and give teachers actual time to experiment will pull ahead.
What Ms. Wang said at that forum in Suzhou stays with me. A museum, not a print shop. That’s the frame I keep coming back to. The tools exist. The question is whether we use them to explore or to produce — to open doors or to shut thinking down.
I think we can do the former. I’ve seen it. The student who caught the fake quotes — she wasn’t just using AI. She was learning to see.
David Jacobson teaches AP World History at Shanghai American School. His students call him Mr. Jacobson.
Sources:
– Thomas Hatch, reporting from Suzhou Experimental Primary School forum (December 2025), via International Education News
– Colleen Kenny, “Content Chaos: Why AI Makes Media Literacy Education’s Most Urgent Task”, LinkedIn
– AmCham China, “Learning in the Age of AI: How China’s International Schools Are Adapting”
– NPR, “In China, AI is no longer optional for some kids. It’s part of the curriculum” (January 2026)
– UNESCO, King Hamad Prize for AI-Driven Creativity and Critical Thinking Projects, 2026
– Frontiers in Education, “Ethical and Regulatory Challenges of Generative AI in Education” (2025)
– Witty School, “How International Schools Are Integrating AI into the 2026 Curriculum”
