🔍 Read the full analysis: What The Microsoft School AI Pact And California’s Laws Say About Oversight on ThorstenMeyerAI.com
Get the latest gadgets 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 U.S. school districts can use a new contract standard for education AI starting November 1, 2026. It restricts data use, monitoring, companion-style features and decisions made without meaningful human review, but it is a contract framework, not a nationwide law. California has separately enacted AI limits for employment and health care, most taking effect in 2027.
Microsoft says U.S. school districts can use a new AI contract standard starting November 1, 2026, setting restrictions on student-data training, monitoring, companion-style features and decisions made without meaningful human review. The agreement is a contract framework, not a federal law; its reach depends on how districts put the terms into agreements with providers. It arrives as California has signed separate laws limiting certain workplace and health-care uses of AI.
The 31-page Memorandum of Agreement was announced September 9 by the American Federation of Teachers (AFT), the United Federation of Teachers (UFT) and Microsoft. The source material identifies the contracting parties as the educator-led National Academy for AI Instruction and Microsoft, with empty signature slots for other AI providers. Microsoft says the protections are available to every U.S. district from November 1. Districts can incorporate the terms into their own contracts, which could give them contractual remedies if a provider violates them.
For covered products—AI designed and marketed for education—the agreement bars using student, teacher or administrator data to train or improve models, subject to a narrow exception for specified safety and security purposes. It also restricts location and behavioral tracking, keystroke logging, passive attention monitoring, biometrics and emotional inference. The text allows exceptions to some listed monitoring restrictions with explicit written approval from the school customer. It bars companion-style features designed to foster emotional attachment or dependency and requires meaningful human review or prior approval for decisions affecting students or teachers.
The agreement also addresses data deletion, breach reporting and automated actions: the source says breaches must be reported within 72 hours, and deleted data removed from active systems within 180 days. Data must remain in the EU or North America. These are terms described in the agreement, not universal legal duties for all education technology. The source says general-purpose productivity and workplace tools are expressly outside its scope, a limit that matters when schools use products not marketed specifically for education.
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 Could Shape School AI
The pact could give school districts a concrete set of terms to request when buying education AI, particularly around student privacy and human oversight. If included in a district contract, the restrictions may be enforceable between the parties; they do not, by themselves, bind every school or AI company in the country. Districts will need to determine which products and data flows fall within the agreement and whether any exceptions have been approved.
For families and educators, the distinction between a contract and a law affects where accountability lies. A district that adopts the terms can negotiate and enforce its agreement, while a district that does not adopt them cannot assume the same protections apply. The restrictions on automated decisions also leave room for AI to assist with school tasks, but call for a person to review or approve covered decisions rather than letting the system decide alone.
California’s statutes take a different route: they establish legal restrictions in specified employment and health-care settings. Alongside a New York law on child-facing chatbots described in the source, the developments point to oversight through several mechanisms—procurement contracts and state legislation—rather than one uniform national AI rule.
Products Worth Considering
Jackery Explorer 2000 v2 Portable Power Station, 2042Wh
Bose QuietComfort True Wireless Noise-Cancelling Earbuds, Black
Apple AirPods Pro 3 Wireless Earbuds with Active Noise Cancellation
As an affiliate, we earn on qualifying purchases.
How the Pact Differs From State Law
The source says AFT and UFT announced the school standard with Microsoft on September 9, 2026. It describes the document as a memorandum of agreement between the National Academy for AI Instruction and Microsoft, not a statute or a general Microsoft policy page. The agreement is designed for districts to place its terms into contracts. AFT President Randi Weingarten argued that protections need to be legally enforceable rather than aspirational; Microsoft President Brad Smith described the standard as setting a high bar for child privacy and AI safety.
California Governor Gavin Newsom signed four measures on September 30, according to the source. AB 1883 addresses AI-based emotion reading and neural data in the workplace; SB 947 limits discipline or firing decided by an algorithm alone; AB 1979 restricts AI from performing health-care functions reserved by law for licensed professionals; and SB 951 requires layoff notices to disclose when AI has replaced jobs. Most are scheduled to take effect January 1, 2027, while SB 947 takes effect July 1, 2027.
The source also reports that New York Governor Kathy Hochul signed a law on October 28 concerning chatbot features for minors and unverified users, including features described as pretending to be human or encouraging emotional dependency. Those state laws have separate scopes and effective dates; they should not be treated as provisions of Microsoft’s school contract.
Who Will Adopt the School Terms?
The source describes Microsoft’s offer as available to every U.S. district but also says the terms become enforceable when a district incorporates them into its own contract. It notes that accounts differ on whether the protections apply automatically or only when a district requests them. That question affects how schools should assess coverage and is not resolved by the material provided.
It is also unclear how many districts will adopt the terms, which other AI providers will sign, and how the agreement will be monitored or disputes handled in practice. The safety exception allows processing for specified risks, but the source does not provide examples of how districts will oversee those cases. The scope exclusion for general-purpose tools may also leave questions when schools use such products for classroom work.
The source summarizes the California and New York laws but does not provide their full statutory text or implementation guidance. Their detailed application, enforcement procedures and any later regulations cannot be established from the material provided. The article’s account of these measures should be read as a summary, not legal advice.
District Contracts and 2027 Start Dates
From November 1, districts can seek to use the Microsoft agreement’s terms in procurement and contract discussions, according to Microsoft. School systems will need to identify which products qualify as education AI, review data practices and decide whether any permitted exceptions should be approved in writing. It remains to be seen how many districts adopt the framework or whether additional providers use its terms.
California’s next milestones are the scheduled effective dates: most of the four measures begin January 1, 2027, and SB 947 begins July 1, 2027. Employers, health-care organizations and software vendors will need to determine how the statutes apply to their systems and practices. New York’s child-chatbot law, signed October 28, is another developing area, but the source does not specify its effective date or implementation schedule.
Key Questions
Is Microsoft’s school AI standard a law?
No. It is a contract framework. Its terms can become enforceable between a district and provider if incorporated into their contract; it is not a nationwide statute.
What does the agreement restrict?
For covered education AI, it restricts model training on school-community data, specified tracking and emotional inference, companion-style features, and decisions made without meaningful human review or prior approval. The source describes a narrow exception for specified safety and security processing.
Does the agreement cover every Microsoft product used at school?
The source says the standard covers AI products designed and marketed for education and expressly excludes general-purpose productivity and workplace tools. A district would need to assess which products fall within that scope.
When do California’s AI laws take effect?
According to the source, most of the four measures signed September 30, 2026, take effect on January 1, 2027. SB 947 is scheduled to take effect on July 1, 2027.
What is still unsettled about the school pact?
The source reports differing descriptions of whether protections apply automatically or only when districts request them. District adoption, participation by other providers and practical monitoring also remain unclear.
Source: ThorstenMeyerAI.com
