Microsoft, the American Federation of Teachers (AFT), and the AFT's New York City affiliate, the United Federation of Teachers (UFT), announced a jointly developed National AI Safety and Privacy Standard on September 9 that gives U.S. school districts contractual recourse over how student and educator data are handled in Microsoft's education-focused AI products2,3.
The standard comprises ten principles covering privacy, security, transparency, and human oversight. Districts can incorporate the protections directly into their existing Microsoft customer agreements, converting what would otherwise be voluntary commitments into enforceable contract terms. A material violation could trigger consequences under the contract, including a process for addressing violations that lets districts terminate agreements and pursue other available actions when problems are not resolved.
Among the specific restrictions: student and educator data, including prompts, AI-generated responses, and uploaded files, cannot be sold, used for unrelated purposes, or used to train general-purpose AI models. The standard also restricts behavioral tracking, long-term profiling, keystroke logging, and continuous attention monitoring. AI systems cannot independently make decisions affecting student discipline, academic placement, or employee evaluation without meaningful human review or prior authorization.
On data deletion, providers generally have 180 days to remove data from active systems following a district's deletion request, subject to district retention policies, legal holds, and other circumstances specified in the agreement. Providers must notify affected education customers within 72 hours after becoming aware of a confirmed or reasonably suspected breach involving covered data. Microsoft must also disclose to families how its tools work in plain language.
Microsoft Vice Chair Brad Smith said the company will extend the agreement to every school district across the country. Other AI providers working with AFT's National Academy for AI Instruction can adopt the standard as well, provided they agree to all of its principles and make equivalent protections available to education customers that request them.
The AFT is the second-largest teachers' union in the United States. In May, AFT President Randi Weingarten called for eliminating screens in kindergarten through second grade, prohibiting student-facing AI in elementary schools, and restricting social companion chatbots for students younger than 16. The Los Angeles Unified School District adopted related approaches to screen and AI use earlier the same week.
The agreement also includes legal protections that hold Microsoft liable for breaches1.
ANALYSIS The standard's enforceability mechanism is its distinguishing feature: by embedding the ten principles in purchase agreements rather than a voluntary pledge, districts gain a contractual lever that most existing ed-tech privacy frameworks lack. Opening the standard to other AI providers through AFT's National Academy for AI Instruction creates a potential template for district-level AI governance that extends beyond a single vendor.