THE ESSENTIALS IN 30 SECONDS

  • The standard, announced on September 9, 2026, applies to Microsoft's US school contracts from November 1 and can be added to existing contracts without renegotiation, according to K-12 Dive.
  • Microsoft commits not to train AI on student or educator data except for narrow safety purposes, and not to sell that data or use it for advertising.
  • AI companions, engagement-maximising features and student tracking are ruled out; the terms are contractually enforceable and include third-party audits.

A set of contractual limits on how Microsoft's AI tools can be used in American schools comes into force on November 1, 2026. The National AI Safety & Privacy Standard was announced on September 9 by the American Federation of Teachers (AFT), the United Federation of Teachers (UFT) and Microsoft, after months of negotiation. From today, it applies to Microsoft's school contracts across the United States.

What changes for schools

According to K-12 Dive, any district in the country can now add the standard to a new or existing Microsoft contract, without renewing or renegotiating it. Under its terms, Microsoft may not use student or educator data to train AI models, except in narrow safety and security cases, and may not sell that data or use it for advertising. Euronews adds that data collected through covered products cannot be used for product development either.

The standard also rules out AI companions and features designed to foster emotional attachment or to keep pupils engaged beyond the learning task. Engadget reports that Microsoft is barred from tracking students and that its AI products cannot make decisions in schools without human oversight. Schools keep control over how their data is used, retained and erased, and Microsoft must explain to families and teachers, in plain language, how its tools work and what they collect.

How it is enforced

The commitments are not a law but contract terms, which the parties describe as legally binding. Microsoft would be liable for breach of contract if it violated them, and the standard calls for third-party audits. Microsoft president Brad Smith said the company would "extend this agreement to every school district across the country."

Limits and open questions

Microsoft is so far the only company to have signed the unions' standard. AFT president Randi Weingarten has said she wants OpenAI and Anthropic to join; Euronews reported that both were in talks with the union, while Google had not said whether it would offer similar terms. Josh Golin of the children's advocacy group Fairplay warned that schools could read strong safeguards as an endorsement of more AI in classrooms, and argued the standard will only matter if other major providers sign on. The protections also depend on districts actually adopting them in their contracts. The deal arrives amid wider pushback on AI and screen time in schools, including temporary AI restrictions in New York City and Los Angeles.