AI use in schools is facing closer scrutiny as districts, parents, and educators ask how student data should be handled. Microsoft has now agreed to a new set of privacy and safety principles aimed at addressing those concerns.
The agreement focuses on student data, AI companions, family transparency, and human oversight.
Microsoft’s New School AI Rules
Microsoft reached the agreement with the American Federation of Teachers (AFT) and its New York City affiliate, the United Federation of Teachers (UFT). The deal includes 10 principles that school districts can make part of their contracts with Microsoft.
One of the main terms prevents Microsoft from using student or educator data to train its AI models. The agreement also calls for limits on the data collected through its education tools.
Families must also receive clear explanations about how the technology works. The language should be easy to understand rather than buried in complicated technical terms.
The rules also prohibit AI companions in school settings. For decisions considered “high-risk,” the agreement requires human review instead of allowing an AI system to make the final call.
The deal comes as some large school districts take a more cautious approach to AI. New York City and Los Angeles both introduced one-year restrictions on AI tools for many students. The temporary bans give officials time to study how these systems should be used and what safeguards schools may need.
Microsoft’s agreement could help districts adopt similar protections without renegotiating entire contracts. Starting in November, school districts can add the new terms to existing or new agreements.
AFT Takes a Careful Approach
The AFT has raised concerns about technology use in classrooms while also supporting teacher training for AI. In May, AFT President Randi Weingarten called for a ban on screens before third grade and “student-facing AI” before middle school.
At the same time, the union announced a $23 million AI training hub for educators funded by Anthropic, Microsoft, and OpenAI.
Weingarten described Microsoft’s new terms as “iron-clad.” She said they were needed because “no one else, including the federal government, has stepped up to do the real work.” She also argued that legally enforceable protections matter more than voluntary promises.
The agreement reflects a growing focus on privacy and accountability as AI enters education. Schools adopting Microsoft’s terms will have clearer requirements around data collection, transparency, AI companions, and human oversight.
Microsoft’s agreement does not settle every question about AI in education. However, it sets specific conditions that districts can place into contracts and gives schools a clearer framework for handling student data and higher-risk AI decisions.
