Why New York City Banned AI in Schools Through 8th Grade

A city decided what its children should keep practicing. Nobody is making that decision for you.

On September 2, New York City announced the broadest restriction on classroom AI in the country. Understanding why they did it gives you a question worth asking about your own week.

What did New York City actually ban?

Mayor Zohran Mamdani and Schools Chancellor Kamar Samuels announced a one-year moratorium on student-facing generative AI for students from 2-K through eighth grade, effective this school year. It covers roughly 600,000 students — about two-thirds of the nation's largest school district.

The specifics matter more than the headline:

• Student-facing generative AI is paused through eighth grade. Software that responds to prompts by producing text, images, audio, or feedback becomes unavailable to those students.
• Companion chatbots are prohibited across all grades, kindergarten through twelfth — including for high schoolers who otherwise retain some access.
• More than 38 previously approved programs will have their AI components discontinued or disabled.
• Screen time is capped at under 45 minutes a day through eighth grade.
• High school students receive two 45-minute AI literacy modules a year, covering how the technology works, privacy, bias, and its likely effect on careers.
• Five supervised pilots remain — Quill, Edia, Brisk Teaching, Playlab, and Intel AI-Ready Schools — with caps of ten to twenty minutes a week under direct teacher supervision.
• Exceptions apply for assistive technology, multilingual learners, and career-readiness programs including computer science.
• Teachers may still use AI for lesson planning. Grading stays human.

Los Angeles Unified has moved to restrict student access to generative AI platforms while it reviews its own safeguards. Other districts are watching.

Why did the city ban it?

This is the part that separates the policy from a simple prohibition.

Officials named what they were protecting: students' critical thinking, and the relationship between a student and a teacher. The restriction follows from that, rather than the other way around.

Mamdani put it plainly: "Just because technology is everywhere doesn't mean it belongs everywhere."

A coalition of educators, parents, students, and experts will spend the year studying the technology's effect and recommending what comes next. The moratorium creates the room for that study to happen.

Why 2023 went differently

New York City banned ChatGPT from school devices and networks in January 2023, then reversed within months. The chancellor at the time described the original ban as a knee-jerk reaction that overlooked what the technology might offer.

Same city. Same technology. Opposite outcome.

The difference lies in the sequence. The 2023 ban was a reaction to a tool that appeared suddenly and looked alarming — a barrier with nothing named behind it. Barriers built that way tend to come down, because when the initial alarm fades there is nothing underneath holding the decision in place.

The 2026 moratorium starts from the thing being protected and works outward. That is a more durable structure, and it explains why the exceptions are so specific. A student with a disability using assistive technology is doing something different from an eighth grader asking a chatbot to explain an equation. Once you know what you are protecting, those distinctions become obvious rather than arbitrary.

I wrote recently about why friction only holds when you know what you're protecting. This is that argument at the scale of 600,000 students, and it is the clearest demonstration of it I have seen.

What are companion chatbots, and why are they banned through high school?

Companion chatbots are applications that present themselves as offering emotional support, friendship, or conversation for its own sake. They are the one category restricted across every grade level, including students who keep limited access to other AI tools.

The reasoning tracks with everything else in the policy. A high schooler using a supervised tool to practice math is building a capacity. A high schooler forming an attachment to software is practicing something else entirely, and the city decided that particular practice can wait.

You can disagree with where they drew that line. What is worth noticing is that they drew it somewhere specific, for a stated reason, rather than either banning everything or permitting everything.

What should adults keep practicing?

Here is where this stops being an education story.

Six hundred thousand children now have a year of protected practice at thinking without assistance. Someone convened a coalition, studied the question, and decided on their behalf what capacities were worth preserving while the evidence comes in.

Nobody is doing that for you.

There is no moratorium on your week. No chancellor deciding which of your capacities should stay in use while the research catches up. The tools sit on your desk, they are genuinely useful, and every decision about what you hand over is yours alone to make, usually in the middle of a busy Tuesday when handing it over is the obvious move.

Which makes the city's question a personal one. What do you want to keep being able to do yourself?

Thinking a problem through before asking for the answer. Writing the difficult message in your own words. Sitting with a decision long enough that your own view surfaces. Noticing what you actually think about something before a summary tells you what to think about it.

Before you automate a piece of your work, the useful question is the same one New York asked: what is this protecting, and what would it cost to stop practicing it?

The Focused Human Lens

Attention behaves like a directional energy. It carries real cost, it organizes what you notice, and it produces coherence when it settles along a single line.

Practice is how that coherence gets built. Working through a hard paragraph, holding a question open, arriving at your own answer — each of those is attention doing the organizing work itself, and each leaves something behind that remains available afterwards.

A tool that supplies the output delivers the result while the organizing work goes elsewhere. The page arrives. The capacity that would have formed stays where it was.

This is why the sequence New York chose matters so much. They identified the capacities worth building first, then decided which tools could stand near them. Direction preceded the barrier. Applied to your own week, the same order produces the same durability: name what you want to keep being able to do, and the decisions about which tools to use answer themselves.

What this makes possible

You get to run this experiment on a scale of one, starting today, without waiting for a coalition to report back.

Pick one capacity you want to keep. Protect the practice of it for a month. Let every other tool stay exactly where it is.

That is a smaller commitment than a city-wide moratorium and it works the same way, because the mechanism is identical: you decided what you were protecting before you decided what to change.

Curious where your attention is going before you start protecting anything? The two-minute attention quiz names the drain costing you the most right now, and gives you one thing to try about it.

A. Karacay is the author of The Focused Human. If you're looking for a weekly practice to help you direct your attention more deliberately, the Weekly Attention Reset Protocol is designed for exactly this. It's free, simple, and built to help you reclaim coherence in a world designed to fragment it.

Stay curious!