🔍 Read the full analysis: The Human Role In Microsoft’s School AI Pact And California’s New Laws on ThorstenMeyerAI.com
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Microsoft says that from Nov. 1, 2026, every U.S. school district can use a contract setting limits on education AI, including restrictions on student-data training, tracking and decisions without human review. The agreement is contractual, not a nationwide law, and its scope is limited to products designed and marketed for education. California has separately enacted workplace and health-care AI restrictions, most taking effect in 2027.
Microsoft says it is making a new AI safety and privacy contract available to every U.S. school district starting Nov. 1, 2026, setting limits on how covered education products use student data and make decisions. The offer comes as California’s new laws set separate boundaries for workplace and health-care AI, putting human review and limits on sensitive data at the center of a broader policy debate.
The 31-page agreement, called the National AI Safety & Privacy Standard, was announced Sept. 9 by the American Federation of Teachers and United Federation of Teachers with Microsoft. The source describes it as a memorandum between Microsoft and the National Academy for AI Instruction, an educator-led training body announced by the AFT in 2025. It includes empty signature spaces for other AI providers. Microsoft says school districts can put its terms into their own contracts; it is not a law or a universal Microsoft policy covering every product.
For products designed and marketed for education, the agreement bars using student, teacher or administrator data to train, fine-tune, benchmark or otherwise improve AI models, subject to a limited safety-related exception. It also prohibits specified forms of tracking and emotion inference, and companion-style features designed to encourage emotional attachment. Covered tools cannot make decisions about students or teachers without meaningful human review or prior approval. The text also requires covered providers to report breaches within 72 hours and delete data from active systems within 180 days.
The restrictions have defined boundaries. The agreement expressly excludes general-purpose productivity and workplace tools, according to the source. The safety exception allows processing data to detect risks such as self-harm, abuse material, grooming, threats, bullying or security attacks, but says that data cannot then be used for general model improvement. The source also says some tracking restrictions may be excepted with a school customer’s explicit written approval. Those qualifications mean the protections depend on which product is covered and how a district contracts for it.
Infographic 01 · The lines drawn this autumn
Seven weeks, one direction: a human stays in charge
From the school contract to state laws: what was signed when, and what the classroom standard actually requires.
- 2 Feb 2025EU: emotion inference banned at work and schoolAI Act Art. 5(1)(f), except medical or safety reasons
- 9 Sep 2026AFT, UFT and Microsoft announce the school standardA contract districts can add to their Microsoft agreements
- 30 Sep 2026California signs AB 1883, SB 947, SB 951, AB 1979Emotion AI at work, robot bosses, AI layoff notices, AI in clinics
- 28 Oct 2026New York signs its chatbot law for minorsS.9051B/A.10379C: no unsafe companion features for under-18s
- 1 Nov 2026Microsoft makes the standard available to all US districtsPer Microsoft’s fact sheet; districts “can add them”
- 1 Jan 2027Most California rules take effectAB 1883, AB 1979, SB 951
- 1 Jul 2027SB 947 becomes operativeHuman corroboration before AI-driven discipline or firing
What Microsoft’s classroom AI may not do
- Learn from students or teachers. No training on their data; de-identification is “not a loophole”.
- Track them. No keystroke logging, attention monitoring, long-term profiling or biometrics.
- Befriend them. No companion features designed to foster emotional attachment.
- Decide alone. No decisions without meaningful human review or prior approval.
Where it is narrower
- Scope: education products only. General-purpose productivity and workplace tools are excluded by definition.
- Opt-in or automatic: Microsoft says districts “can add” the terms; the agreement says “upon request within 90 days”.
- Exceptions: a narrow safety exception for training; the no-tracking list allows written approval by the school.
“Anything less than legally enforceable provisions is simply a wish list.”Randi Weingarten, AFT President, 9 September 2026
Sources: AFT/UFT/Microsoft release, 9 Sep 2026 · Standard MOA v1.0 · Microsoft fact sheet · Governor of California, 30 Sep 2026 · Governor of New York, 28 Oct 2026 · EU AI Act, Art. 5(1)(f). Effective dates for California follow the chaptered texts and the state’s default start date for regular-session bills.
Powered by Thorsten Meyer AI
Where Human Review Becomes a Requirement
The shared policy question is not whether schools, employers or clinicians may use AI at all, but which decisions remain under human control. For families, the school agreement addresses the handling of children’s data and limits AI features that could monitor behavior or imitate a personal relationship. For educators, it sets contractual expectations around review rather than allowing covered tools to make consequential decisions on their own.
For districts, the practical force comes from procurement: if a school puts the terms in a contract, the source says it can end that contract and seek damages if the terms are breached. That makes the agreement different from a voluntary aspiration, though it does not bind every school or AI provider automatically. The source quotes AFT President Randi Weingarten saying that “anything less than legally enforceable provisions is simply a wish list.”
California’s laws move some similar concerns into state legislation, but in different settings and on separate timelines. Taken together, the developments could prompt organizations to check whether their software reads mood or attention, influences discipline or firing, or handles decisions that law reserves for people with professional qualifications. They do not amount to a single nationwide AI rule, and the contract’s reach should not be confused with the reach of state law.
Popular right now
Bose QuietComfort True Wireless Noise-Cancelling Earbuds, Black
Shark Rocket Ultra-Light Corded Stick Vacuum, HV301, Gray/Orange
Dyson Car+Boat Cordless Handheld Vacuum
As an affiliate, we earn on qualifying purchases.
Three Policy Moves, Different Reach
The school agreement followed a Sept. 9 announcement by the AFT, UFT and Microsoft. Microsoft’s stated start date for making the terms available to U.S. districts is Nov. 1, 2026. Its contractual approach lets each district decide whether and how to incorporate the terms into its agreement with Microsoft. The source notes that descriptions differ on whether the provisions apply automatically or only when a district requests them, making the contract language and procurement process relevant to schools.
On Sept. 30, California Gov. Gavin Newsom signed four measures identified in the source as AB 1883, SB 947, AB 1979 and SB 951. They address emotion or neural-data collection at work, algorithm-only firing or discipline, the boundary between health-care AI and work reserved to licensed professionals, and notices when AI replaces jobs. The source says most take effect Jan. 1, 2027, while SB 947 takes effect July 1, 2027.
New York added a separate measure on Oct. 28, when Gov. Kathy Hochul signed a law concerning chatbots used by minors and unverified users. The source describes restrictions on unsafe companion features, including chatbots that present themselves as human or foster emotional dependency. These are distinct developments: the school pact is a contractual standard, while the California and New York measures are laws with their own scope and effective dates.
““Anything less than legally enforceable provisions is simply a wish list.””
— Randi Weingarten, president of the American Federation of Teachers
Contract Coverage and Enforcement
It is not yet clear how many districts will adopt the agreement, whether other AI providers will sign it, or how consistently its protections will appear in local contracts. The source describes the district’s ability to terminate a contract and seek damages, but does not provide details about a dispute process or any enforcement actions. It also reports differing descriptions of whether the standard applies automatically or only when a district asks for it.
The agreement’s exclusions and exceptions leave practical questions. Its stated scope does not cover general-purpose productivity and workplace products, and some tracking restrictions can be waived through a school’s written approval. The source does not identify which products Microsoft will treat as covered education products or explain how districts will distinguish them from other tools. The impact of California’s laws will likewise depend on their application and enforcement after their effective dates.
District Procurement and 2027 Deadlines
From Nov. 1, districts can consider whether to place the Microsoft agreement’s terms in their own contracts. School procurement teams may need to identify covered products, review data practices and decide whether any permitted exceptions fit their policies. Whether the academy’s training and the contract provisions change day-to-day classroom practices remains to be seen.
California’s next milestones are the Jan. 1, 2027, effective dates for most of the four laws and the July 1, 2027, date for SB 947, according to the source. Employers and health-care organizations will need to determine how the provisions apply to their systems and decisions. New York’s signed chatbot law adds another set of requirements for services used by children. Further guidance, implementation details and any legal challenges were not provided in the source.
Key Questions
Is Microsoft’s school AI standard a law?
No. It is a contractual agreement. A district can choose to put its terms into its Microsoft contract; it is not a law that automatically applies to every school or AI provider.
What does the agreement prohibit?
For covered education AI products, it restricts using student and educator data to improve AI models, specified tracking and emotion inference, and companion features designed to foster emotional attachment. It also requires human review or prior approval for covered decisions about students or teachers.
Does the agreement cover every Microsoft AI tool used at school?
Not according to the source. It covers products designed and marketed for education and expressly excludes general-purpose productivity and workplace tools. Districts will need to establish which products fall under their contracts.
When do California’s AI laws take effect?
The source says most of the four laws signed Sept. 30 take effect on Jan. 1, 2027. SB 947, concerning discipline or firing decided by an algorithm alone, takes effect July 1, 2027.
Source: ThorstenMeyerAI.com
