New York City just pulled the plug on generative AI in elementary and middle schools, imposing a one-year moratorium that affects roughly 600,000 students. The ban covers any student-facing software that uses generative AI, according to reporting from Politico and the AP. It’s a sharp reversal for a school system that, like many districts, had been cautiously experimenting with AI tools over the past two years. The move signals that concerns about children’s data privacy, algorithmic bias, and the developmental impact of chatbots on young learners have finally outweighed the promise of personalized tutoring.
This isn’t happening in a vacuum. The same anxieties are driving a parallel conversation about child-monitoring apps, which scan kids’ texts, photos, and emails for signs of danger. Those tools have prevented real harm—suicide attempts, predators, school shootings—but they’ve also generated false alarms, eroded trust between parents and children, and added new layers of anxiety to family life. Researchers in the child-safety space argue that better approaches exist, though the article doesn’t spell out exactly what those look like. The common thread is clear: adults are increasingly uncomfortable with algorithms making decisions about children’s wellbeing.
The broader context makes NYC’s decision look less like a Luddite overreaction and more like a course correction. Just this week, the Justice Department sided with OpenAI in its copyright fight against the New York Times, arguing that AI training can constitute fair use. Meanwhile, the Trump administration is urging G20 nations to let AI models train on creators’ work. The message from Washington is that AI should move faster. New York City’s schools just said: not for our kids.
Whether a one-year moratorium actually accomplishes anything is an open question. Generative AI isn’t going away, and students will encounter these tools regardless of what their schools officially permit. A ban might simply push usage underground, where there’s even less oversight. But for a school system still recovering from pandemic learning loss, hitting pause to figure out what student-facing AI actually does to developing minds may be the most sensible tech policy in circulation right now.
Elsewhere, Taiwan uncovered a Chinese network targeting its chip sector with 166 linked cases of illegal operations and talent poaching. Uber launched the UK’s first commercial robotaxi service using technology from startup Wayve. Google dodged a forced breakup of its ad tech business after a US judge rejected the government’s demand, though the company must change how it treats rivals. And in a detail that feels almost too on-the-nose for this week’s AI news: generative AI food images have started appearing on real restaurant menus, and they frequently look revolting.
