For the last two weeks, education news feed was inundated with the news on New York City banning AI for students through eighth grade, and Los Angeles Unified with almost no public notice restricting AI for everyone. I read both stories more than once, and I didn’t agree with them either on the substance, nor the process, and don’t support the message it sends to the kids sitting in those classrooms this fall.
AI is not a disruption our school districts can wait to pass. It’s already a tool that growing share of adults use at work, and it will be even more entrenched by the time this year’s freshmen graduate.
US is behind on building the safeguards that should exist in schools and everywhere else. But being behind on safeguards is an argument for building them faster. It is not an argument for pretending the technology isn’t there. The next generation has to be ready for an AI world regardless of if we have finished writing the rules for it, and the districts that treat that as someone else’s problem are the ones I worry about most.
I decided to try to find school districts in California which decided to incorporate AI in their education process and what does it mean to their students and create a compounded profile of students in those school districts and their experiences.
One goes to school in Los Angeles Unified. One goes to school forty minutes south, in Anaheim Union High School District. I’ll call them Maya and Diego. They are composites because those personas are built from the policies and systems each district adopted. A decade from now, I think their lives look different because of a decision neither of those students or their families and supporters made.
Maya, in Los Angeles
Maya starts her junior year this month carrying a district-issued Chromebook that cannot open a generative AI tool of any kind. She will not be able to draft an essay online; check her math reasoning or ask a clarifying question about a reading she didn’t understand. It applies to every grade, on every district device, for the whole 2026-27 school year.
Before this year, Maya could have used approved AI tools once she turned 13 and finished a digital citizenship course. Now that access is gone while the district figures out what comes next. LAUSD says it still wants Maya to be AI-literate. It just hasn’t said how she’s supposed to get there without touching the thing she’s meant to become literate in.
So Maya will do what plenty of students already do when a tool is banned at school and ubiquitous everywhere else: she’ll use it anyway, on her own phone, without a teacher in the room to show her what good and bad use looks like. The ban doesn’t protect her from AI. It just removes the adults from the room while she learns to use it badly, alone, for a tool she’ll be expected to use fluently at any job she takes after graduation.
Diego, in Anaheim
Diego starts his junior year forty minutes south, and his school doesn’t measure him primarily by a standardized test. AUHSD replaced that signal years ago with a graduate profile built around five competencies such as creativity, critical thinking, collaboration, communication, and character. Diego’s biggest assignment this semester isn’t a bubble sheet. It’s a public “soapbox” speech and a capstone project, graded against those five Cs by a teacher who knows his name.
An AI tool, built through a research partnership between AUHSD and UC Irvine and funded by a $1.1 million Chan Zuckerberg Initiative grant, helps Diego’s teacher read his reflections and project work at a depth a single grade could never capture. It helps him to track growth in his voice and sense of purpose across the year, not just whether he picked the right answer. Diego’s teachers went through real training to get here. Some veteran staff resisted both the new grading model and the AI support behind it. Superintendent Michael Matsuda has said as much publicly and the district responded with more training, not a retreat.
Diego uses AI the way he’ll use it in almost any career he chooses: as an instrument, in a room with adults who show him where it helps and where it lies to him.
The same tool, two different systems
Maya and Diego are twelve miles and one county line apart. Neither of them chose their district’s philosophy of education. Neither of them voted on a screen-time policy or sat in a superintendent’s cabinet meeting. And yet one of them is being trained to think alongside AI under adult supervision, and one of them is being told to pretend it doesn’t exist until further notice.
I don’t think this is really a story about AI. I think it’s a story about what each district believes school is for. Anaheim built its answer first and put creativity, critical thinking, collaboration, communication, character in the center. Then it asked which tools serve that purpose. AI passed the test because Anaheim already knew what it was testing for. LAUSD, as far as I can tell, hasn’t finished answering the purpose question, so it reached for the only lever available on short notice: restrict access, review later, hope the review produces clarity the initial decision didn’t have.
Maya will graduate having been told, for one formative year, that the most consequential technology of her lifetime was something to be kept out of her hands rather than something to be taught. Diego will graduate having practiced, under supervision, exactly the kind of judgment his district decided mattered most.
Neither of these students asked for this experiment. Every district running one right now is making that choice on behalf of kids who won’t get the year back. I’d rather be the superintendent who can explain, ten years from now, why I built toward something than the one who can only explain what I restricted while I figured it out.
Leave a Reply