The nation's two largest school districts are pausing student-facing generative artificial intelligence as the school year begins. While New York City and Los Angeles schools establish stricter boundaries, major technology providers face growing pressure to legally guarantee student data privacy. Schools are moving away from rapid classroom AI adoption toward a more cautious, regulated approach.
What Happened
New York City public schools launched a one-year moratorium on generative AI tools for approximately 600,000 students in grades 3-K through eighth, according to the NYC Mayor's Office. The city has disabled generative AI features in more than 38 educational software programs that were previously allowed in classrooms, according to The American Quorum. High school students still have supervised access but must first complete two 45-minute AI-literacy training modules covering bias, privacy, and proper usage, as reported by The American Quorum.
Similarly, the Los Angeles Unified School District has implemented a districtwide moratorium on student generative AI tools on school-owned devices, as detailed by The American Quorum. Both districts have made exceptions. Teachers may still use AI tools for administrative planning, and students with disabilities or language-learning needs retain access to specialized assistive technologies.
The Bigger Picture
Local school district policies are shifting as major software providers change their own rules. As we previously reported, Microsoft signed a legally binding agreement with the American Federation of Teachers (AFT) to establish a "National AI Safety & Privacy Standard" for schools. Under this 30-page memorandum, described by Education Week, Microsoft contractually promises not to use student or teacher data to train its AI models and will block AI features designed to mimic emotional companions.
Google has also set up similar protections. The company designated Gemini for Education and Gemini Notebook as "core services" under its Google Workspace Terms of Service, putting them under strict data protection policies. Google's Gemini for Education recently earned the Common Sense Privacy Seal. This certification confirms that Google does not train its AI on student data. It also prevents human review of classroom prompts and keeps all content within the school district's secure domain, as detailed by Common Sense Privacy.
What This Means for Families
For parents and educators, these decisions focus attention on data security. The AFT’s push for safety standards, including a ban on social companion chatbots for users under 16, shows a growing concern over how conversational AI affects children emotionally.
School districts now have more bargaining power. Instead of accepting standard commercial terms, school boards can require technology vendors to sign agreements modeled after the AFT-Microsoft standards. Because policies vary widely by district, parents must track which tools their children's schools approve.
What You Can Do
First, contact your school principal or district technology director to ask if classroom AI providers have signed legally binding agreements that prohibit using student data for model training.
Second, if you live in a district with an active moratorium, verify which specialized high school programs or assistive technology tools are exempt so your child has the necessary access.
Finally, look for platforms that carry third-party safety credentials, such as the Common Sense Privacy Seal, which confirm that student work remains secure and private.