🔍 Read the full analysis: The Rules For Human Control In School AI: Microsoft’s Pact And California’s New Laws on ThorstenMeyerAI.com
Get privacy and security gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Microsoft says a school AI agreement developed with educator groups will be available to U.S. school districts from Nov. 1, 2026. The contract terms restrict training on student and educator data, tracking and companion-style features, and require meaningful human review of covered decisions. The agreement is contractual, not a new law, and applies to education-designed products rather than general-purpose tools.
Microsoft says that from Nov. 1, 2026, every U.S. school district can use contract terms restricting how covered classroom AI handles student and educator data and decisions. The agreement, developed with educator groups, bars training on that data, prohibits specified tracking and companion-style features, and requires meaningful human review for decisions about students or teachers; it is a contract framework, not a federal law.
The standard is a 31-page Memorandum of Agreement between Microsoft and the National Academy for AI Instruction, an educator-led training body that the American Federation of Teachers (AFT) announced in 2025. The document includes empty signature slots for other AI providers. School districts can put its terms into their own Microsoft contracts, according to the source report. That means enforcement would depend on the district’s contract, rather than the agreement operating as a standalone statute.
Under the terms described in the agreement, providers may not use prompts, answers, uploaded files, behavioral signals or metadata from students, teachers and administrators to train, fine-tune, benchmark or otherwise improve AI models. The restriction is described as continuing after a contract ends. A narrow exception permits data processing for specified safety concerns, including self-harm risk, abuse material, grooming, threats, bullying and security attacks; that data cannot then be used for general model improvement.
The agreement also bars specified forms of location and behavioral tracking, continuous passive attention monitoring, keystroke logging and biometric collection, including emotional inference. It prohibits companion-style features designed to foster emotional attachment or dependency. Covered systems must not make decisions about students or teachers without meaningful human review or prior approval. The text allows some listed tracking exceptions with a school customer’s explicit written approval, so the restrictions are not framed as exception-free.
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
Contract Terms Reach School Procurement
The agreement gives districts a potential way to set privacy and human-review conditions when buying covered AI products. That matters because the standard is not itself a law: its practical force comes from a school district choosing to incorporate terms into a contract and then using the contractual remedies available if a provider breaches them. Districts will need to examine the final agreement and their own contracts to determine what protections apply.
For families and educators, the restrictions address concerns that AI could reuse classroom conversations to improve models, monitor students, imitate a personal relationship or make consequential decisions without a person’s review. For providers, the terms could affect product design and procurement conversations. The agreement’s reach is limited, however: the source says it covers products designed and marketed for education, while expressly excluding general-purpose productivity and workplace tools.
The standard sits alongside separate state-level developments described in the source. California signed laws concerning workplace AI and health care in September, while New York signed a law concerning chatbot interactions with children in October. These are distinct measures with different scopes and start dates; they do not turn Microsoft’s school agreement into a law or establish one uniform national rule.
Products Worth Considering
Logitech MX Mechanical Wireless Illuminated Keyboard, Tactile Quiet, Graphite
Samsung Galaxy Tab S10 FE 128GB Wi-Fi Android Tablet, Blue
Bose QuietComfort True Wireless Noise-Cancelling Earbuds, Black
As an affiliate, we earn on qualifying purchases.
Three Separate Routes to Human Oversight
The school agreement was announced on Sept. 9 by the AFT and United Federation of Teachers with Microsoft, according to the source. The AFT’s Randi Weingarten said enforceable provisions mattered because, in her words, “anything less than legally enforceable provisions is simply a wish list.” Microsoft President Brad Smith described the standard as setting “a high bar for child privacy and AI safety.” Those are the parties’ characterizations of the agreement, not independent findings about its effects.
California’s Governor Gavin Newsom signed four measures on Sept. 30, the source reports. AB 1883 addresses AI that reads workers’ emotions or collects neural data; SB 947 restricts firing or discipline decided by an algorithm alone; AB 1979 limits AI in health care where the work is reserved for licensed professionals; and SB 951 requires layoff notices to say when AI replaced jobs. Most are scheduled to take effect Jan. 1, 2027, while SB 947 is scheduled for July 1, 2027.
New York Governor Kathy Hochul signed a children’s chatbot law on Oct. 28, according to the same source. It restricts described unsafe companion features for minors and unverified users, including chatbots that pretend to be human or encourage emotional dependency. The measures share a focus on limits to automated systems, but their legal mechanisms, covered users and obligations differ.
““Anything less than legally enforceable provisions is simply a wish list.””
— Randi Weingarten, AFT president
Adoption and Scope Still Need Clarity
It is not clear from the source whether the terms apply automatically to Microsoft’s education AI products or only when a district asks to include them. The source says Microsoft describes availability differently from the agreement and press reports. Districts and providers will need to clarify how the standard is incorporated into individual contracts and which products qualify as designed and marketed for education.
The extent of adoption, enforcement and compliance is also unknown. The source describes possible contract termination and damages as remedies, but does not provide details of any district’s signed contract or explain how disputes would be handled. The agreement’s safety exception and its allowance for school-approved exceptions to some tracking restrictions also leave implementation questions. No evidence is provided about whether the restrictions have already changed products or reduced privacy risks.
The California and New York measures have separate statutory scopes and timelines. The source does not provide the full statutory text or implementation guidance, so readers should not treat its brief descriptions as a complete account of each law. The laws’ practical effects will depend on their final interpretation and enforcement.
District Contracts Set the Next Milestone
Starting Nov. 1, 2026, districts can seek to use the Microsoft agreement’s terms in procurement and contract discussions. The next concrete signals will be whether districts adopt the provisions, which products are covered, and how providers document compliance. Districts will also have to determine whether their contracts address the agreement’s data-retention, breach-reporting and data-location requirements, described in the source as a 72-hour breach notice, deletion from active systems within 180 days and storage in the EU or North America.
California’s laws are scheduled to begin taking effect in January and July 2027, depending on the measure. New York’s signed chatbot law will proceed through its own implementation and enforcement process. Until more details emerge, the confirmed development is a set of separate contractual and legal rules—not a single AI standard governing every classroom, workplace or chatbot.
Key Questions
Is Microsoft’s school AI standard a law?
No. It is a contract agreement. Its enforceability for a district depends on whether the district incorporates the terms into its contract and on the remedies available under that contract.
What student data may covered AI use to train models?
The agreement described in the source bars using student and educator prompts, responses, uploaded files, behavioral signals and metadata to train or improve AI models. It includes a limited safety-processing exception, but says the data cannot then flow into general model improvement.
Does the agreement cover every Microsoft AI tool used at school?
The source says it covers products designed and marketed for education and expressly excludes general-purpose productivity and workplace tools. How districts and Microsoft determine coverage in individual contracts remains unclear.
When do California’s AI laws take effect?
According to the source, most of the four signed measures are scheduled to take effect on Jan. 1, 2027. SB 947 is scheduled to take effect on July 1, 2027.
They set different limits on what AI may do and where human oversight is expected. Their legal forms, covered settings and requirements are not identical, and Microsoft’s agreement is contractual rather than a law.
Source: ThorstenMeyerAI.com
