AI privacy deal

Teachers union reaches AI privacy deal with Microsoft

The tech giant has agreed not to use student or teacher data to train AI and not to track students using the technology.

By Madina Touré

The American Federation of Teachers and Microsoft have reached an agreement establishing privacy and safety protections for the use of artificial intelligence in schools.

The AI privacy standard consists of 10 legally enforceable protections, including an agreement by the tech giant to not use student or teacher data to train or improve an AI model, with the exception of “narrow safety and security” circumstances, according to a Microsoft fact sheet shared with POLITICO.

The standard bars Microsoft from tracking students and prohibits its AI from making decisions in schools without human oversight. It also requires the company to be transparent with educators and parents on how its tools work.

AFT, the nation’s second-largest teachers union, described the “first of its kind” agreement as an effort to fill a void left by the absence of meaningful federal and state regulations on AI in schools. Efforts to pass federal legislation to establish AI guardrails in schools have not come to fruition.

“We have forged a hard-fought, iron-clad privacy agreement with real teeth that protects students and families, because no one else, including the federal government, has stepped up to do the real work,” AFT President Randi Weingarten said in a statement. “We can get angrier and angrier, or we can act decisively; anything less than legally enforceable provisions is simply a wish list.”

If Microsoft violates the agreement, it is liable for breach of contract, according to the union.

The announcement represents the latest push to clamp down on and regulate the presence of AI in schools, amid growing bipartisan political backlash against Big Tech across the country. It also comes as a nationwide movement to limit screen time for young children takes hold.

The agreement is the result of months of negotiations. An AFT spokesperson said the union is in active talks with other tech companies — OpenAI and Anthropic — to adopt the standard’s provisions.

Microsoft plans to make the protections available to every school district in the U.S. beginning on Nov. 1, according to the fact sheet. Districts can add them to new or existing customer agreements with Microsoft without renegotiating or renewing their contracts.

The tech giant also agreed to give education leaders and families “clear, plain-language information” on how its AI products work in the classroom, the controls schools have over them and safeguards for students, the fact sheet states.

“This standard sets a high bar for child privacy and AI safety, and we’ll extend this agreement to every school district across the country,” said Brad Smith, Microsoft vice chair and president, in a statement.

AFT, along with Microsoft and the United Federation of Teachers — New York City’s powerful teachers union — will announce the agreement at a press conference Wednesday morning. The agreement is between AFT’s National Academy for AI Instruction and Microsoft, and the UFT is a partner in that institute.

Michael Mulgrew, president of the 200,000-member United Federation of Teachers, said the agreement begins to level the playing field in a “competitive, ever-changing arena” that lacks guardrails for children and school communities.

“Educators are stepping in to do what we do best: protect our students,” Mulgrew said in a statement. “This agreement gives families and school districts the information and privacy protections they have been demanding, and it empowers school districts to end agreements and seek damages from bad actors who break the rules.”

The New York City school system, the nation’s largest, announced a one-year moratorium on AI in schools for students in 2K through eighth grade earlier this month, along with recommended screen time limits for younger children.

Hours later, Los Angeles Unified School District, the nation’s second-largest school system, went further — revealing that it is banning AI for all students.

Weingarten said the agreement between the teachers union and Microsoft “adds to the momentum around online safety for kids and educators” in the wake of the city adopting AFT’s framework to ban screens and AI use for younger children.

Last year, the AFT announced that it was launching the National Academy for AI Instruction to provide free AI training and curriculum for AFT members, supported by $23 million from Microsoft, Anthropic and OpenAI.

The academy drew backlash from some teachers, who argued that it could erode teachers’ autonomy over how AI is used in the classroom. At the time, Weingarten defended the partnership, saying the union was working out privacy and safety standards and that the tech giants signaled their support for that.

In a May address, Weingarten called for a ban on screens — including online tests — for students in prekindergarten through second grade and an end to “student-facing” AI in elementary schools. That was part of a separate 10-point plan that did not propose an AI ban but rather call for balancing the benefits of technology “while mitigating the harms.”

Weingarten also said she supports exceptions for compelling cases, such as students with disabilities required to use screens as part of their services.