A New York City public middle and high school for girls is using ChatGPT-generated lesson plans to help students examine the chatbot's usefulness and pitfalls
or more important than—knowing how to program computers.” fascinating https://twitter.com/... Fabio Chiusi / @fabiochiusi : “Ms. Shuman was using the algorithm-generated lesson to examine the chatbot's potential usefulness and pitfalls with her students” https://www.nytimes.com/... Natasha Singer / @natashanyt : What happens when you ask student programmers to critique ChatGPT and other A.I. technologies? They have lots of sharp insights, natch. https://www.nytimes.com/... H. James Wilson / @hjameswilson : Great idea and much needed these days. Some schools are asking student programmers to think critically about rapid advances in artificial intelligence. #AI #futureofwork #futureofeducation @AccentureRSRCH @pauldaugh https://www.nytimes.com/...
Context & Ripple Effects
In February 2023, when most of New York City's public school system still treated ChatGPT as something to block, one public middle-and-high school for girls went the opposite way: teacher Ms. Shuman used an algorithm-generated lesson plan itself as the text, having student programmers dissect what the chatbot gets right and where it fails. The approach anticipated where the system was heading — by May, New York City Public Schools lifted its ChatGPT ban outright, promising support for educators and students exploring AI.
First-order effects
- Shuman's students get direct experience stress-testing AI output instead of merely consuming it — mirroring what a Wharton professor found when his classes were required to write essays with ChatGPT: students quickly surfaced accuracy and bias issues on their own.
Second-order effects
- The district-level reversal validates this classroom-first approach and pushes other cautious educators to follow, even as some question whether generative AI can enter classrooms without undermining numeracy and literacy.
Third-order effects
- Lesson content is becoming a contested market: universities, tech companies, and nonprofits now sell ready-made AI curricula, and OpenAI has since shipped a free online course for K-12 teachers — meaning the tool being critiqued is also funding the training on how to critique it.
The trend: US schools are pivoting from banning ChatGPT to building AI literacy around it, with early adopter classrooms like this one setting the template the district later formalized.