Current status: Last reviewed August 17, 2026. This is a primary-source implementation tracker, not a legal opinion or a complete digest of every state rule.
As of August 17, 2026, there is no single state-law pattern for school AI. Ohio, Tennessee, Utah, North Carolina, Oklahoma, and Maryland impose different local policy duties or deadlines. Oklahoma also creates a parent opt-out from student-facing AI tools and requires alternative access to core instruction. Texas applies public-sector AI governance duties to school districts as local governments. California and Illinois direct state-level policy, guidance, curriculum, or standards work without creating the same blanket local mandate. Massachusetts, Minnesota, Oregon, and Washington publish official guidance for local implementation.
Last reviewed August 17, 2026. This is a primary-source implementation tracker, not a legal opinion or a complete digest of every state rule. A state omitted from the detailed table has not been determined to have no relevant law or guidance. State privacy, student-records, procurement, civil-rights, accessibility, cybersecurity, synthetic-media, assessment, and local board rules can apply even when an AI-specific mandate is absent.
Compact state comparison
| State | Classification | Verified effect | Immediate district action |
|---|---|---|---|
| California | Binding state duties plus nonbinding model policy | AB 2876 directs curriculum-framework work; SB 1288 required state guidance and a model policy, which CDE labels exemplary and nonmandatory | Distinguish state duties from local recommendations; adapt the model only after local review |
| Illinois | Binding state-board duty plus state guidance | Public Act 104-0399 required statewide guidance and continuing updates; the cited law does not itself impose the same blanket local policy deadline | Review current ISBE guidance and decide which local policy, training, and procurement changes to adopt |
| Maryland | Binding law, local policy and coordinator duties | Local systems must align a policy within 120 days after state guidance and designate a central-office AI coordinator | Identify the coordinator now; track guidance publication and the resulting local deadline |
| Ohio | Binding law, local policy duty | Traditional public school districts, community schools, and STEM schools were required to adopt an AI-use policy by July 1, 2026 | Confirm adoption and implementation records; use the state model as an optional benchmark |
| Tennessee | Binding law, local policy and training duties | Local boards and public charter governing bodies must maintain and report AI-use policies; grade 6-12 teacher AI professional development has a future completion requirement | Confirm annual policy reporting and enforcement documentation; plan approved professional development |
| Utah | Binding law, state model and local policy duties | State model policies, local adoption, public process, training, parent communication, and review requirements phase in under the 2026 law | Map the December 2026 state-model milestone and July 2027 local-policy deadline |
| Texas | Binding law, public-sector AI governance | School districts, as local governments, must adopt the state AI ethics code and heightened-scrutiny standards, review covered deployments, and use required notices where applicable | Inventory public-facing and consequential-decision AI; map each use to the final DIR standards and templates |
| Massachusetts | State education guidance | 2026 guidance supports districts that choose to explore or implement AI and explicitly leaves the decision with local leaders | Use the implementation framework without presenting it as a statewide AI-use mandate |
| Minnesota | State education guidance | MDE’s 2026 guidance emphasizes people-centered use, privacy, tool evaluation, local governance, and AI as a learning aid rather than a replacement | Use the guidance to structure local evaluation, privacy review, instructional boundaries, and professional learning |
| North Carolina | Binding law, state model, local policy, and training duties | State model due December 31, 2026; public-school-unit policies due June 30, 2027; teacher training due June 30, 2028 | Assign owners for model review, policy adoption, training completion, and standards updates |
| Oklahoma | Binding law, student-facing safeguards and local policy duty | Effective July 1, 2026; requires human review, limits high-stakes use, annual disclosure, parent opt-out, alternative instruction, and a local policy before 2027-28 | Build a tool inventory, parent-choice workflow, alternative instruction, and board-policy implementation record |
| Oregon | State education guidance | ODE’s current web guidance recommends system-level planning, local policy, tool review, training, and family engagement | Use the flexible guidance to review local policy and protocols |
| Washington | State education guidance | OSPI publishes a human-centered framework, responsible-use guidance, and an LEA leadership checklist | Use the framework to inform local policy, stakeholder work, and implementation |
The classification describes the cited source, not every legal obligation in the state.
How to read a state AI requirement
For every source, record six facts:
- Legal status: enacted statute, effective code, administrative rule, standard, agency guidance, model policy, or pending bill.
- Covered entity: state board, education agency, school district, charter school, educator, student, vendor, or another actor.
- Required action: adopt, report, train, notify, review, prohibit, consider, or provide guidance.
- Date: enactment, effective date, implementation deadline, recurring report date, and review date.
- Scope: instruction, assignments, student-facing tools, curriculum standards, grading, data, professional development, or another defined activity.
- Official source: current code, enacted chapter, rule register, or agency publication.
Do not turn “shall consider” into “shall teach.” Do not turn agency recommendations into a statute. Do not treat an introduced bill as enacted law.
Binding requirements verified in this review
California
Classification: Binding state-level duties plus a nonbinding state model policy.
California AB 2876 was chaptered in 2024 and amended Education Code Section 33548. It defines AI literacy and requires the Instructional Quality Commission to consider incorporating AI literacy into the mathematics, science, and history-social science curriculum frameworks when those frameworks are next revised after January 1, 2025. It also requires the commission to consider AI literacy in criteria for evaluating instructional materials when the State Board of Education next adopts materials in those subjects after January 1, 2025.
SB 1288 separately required the California Department of Education to issue AI guidance by January 1, 2026 and, with working-group input, a model policy by July 1, 2026. CDE’s published model addresses authorized tools, privacy-by-design review, human oversight, AI-assisted grading and feedback, vendor evaluation, monitoring, and autonomous student-facing AI. The department expressly labels the model exemplary and nonmandatory under Education Code Section 33308.5.
These laws act at the state framework, guidance, and model-policy levels. They should not be summarized as a blanket mandate that every district adopt the model, offer a standalone AI course, or implement every recommended consent practice.
District implementation questions:
- Which affected state frameworks or adoption criteria are currently under revision?
- How will future AI-literacy content map to local curriculum and digital citizenship work?
- Which recommendations in the state model should be adopted, modified, or rejected through local review?
- If a district permits autonomous student-facing AI, how will it handle the model policy’s recommended annual guardian-consent process and non-AI alternative?
Official source: California AB 2876, Chapter 927
Official source: California SB 1288, Chapter 893
Official source: California State Model Policy on the Use of AI in Education
California implementation guide: SB 1288 and the state model policy
Illinois
Classification: Binding state-board duty plus official state guidance.
Illinois Public Act 104-0399 required the State Board of Education to develop statewide guidance for school districts and educators by July 1, 2026 and to update it continuously. The law focuses on the state board’s work. It does not, by itself, create the same blanket local adoption deadline found in states such as Ohio or Oklahoma.
ISBE’s published guidance is a practical local reference for responsible, context-sensitive use across prekindergarten through grade 12. District teams should still separate recommendations in that guidance from duties created elsewhere in Illinois law, regulation, procurement rules, privacy obligations, or local board policy.
District implementation questions:
- Which local owners will review future ISBE updates and document resulting changes?
- Which uses require additional privacy, procurement, accessibility, civil-rights, or human-review controls?
- How will educator guidance distinguish approved instructional assistance from decisions that remain human-owned?
Official source: Illinois Public Act 104-0399
Official source: Illinois State Board of Education AI Guidance
Maryland
Classification: Binding law, state guidance and rubric duties, local policy and coordinator duties.
Maryland SB 720 became Chapter 634 and took effect June 1, 2026. It directs the Maryland State Department of Education to publish guidance and an AI-tool evaluation rubric. Each local school system must develop and implement a policy aligned with that guidance within 120 days after the guidance is released and designate a central-office, noninstructional staff member as its AI coordinator.
The law also addresses procurement alignment, AI literacy in K-12 standards by June 1, 2027, and a professional-development program with implementation milestones. The exact local-policy due date therefore depends on the official guidance-release date, not merely the statute’s effective date.
District implementation questions:
- Who is the designated AI coordinator, and what authority and reporting line does that role have?
- What event starts the 120-day local-policy clock, and who records that date?
- How will the state rubric connect to local procurement, approved-tool records, training, and policy enforcement?
Official source: Maryland SB 720, Chapter 634
Maryland implementation guide: AI Ready Schools Act requirements
Ohio
Classification: Binding law, local AI-use policy duty.
Ohio Revised Code Section 3301.24, enacted through House Bill 96 and effective September 30, 2025, required each school district, community school, and STEM school to adopt a policy on the use of artificial intelligence by July 1, 2026. The Ohio Department of Education and Workforce published a model policy that a covered district or school may adopt or adapt. The model is an option, while adoption of a local policy is the statutory duty.
District implementation questions:
- Was the policy adopted by the covered district or school by the statutory deadline?
- Does the adoption record identify the adopted version and its effective date?
- Who owns implementation, staff communication, training, and periodic review?
- If the state model was adapted, which local provisions differ and why?
Official source: Ohio Revised Code Section 3301.24
Official source: Ohio AI Model Policy for Districts and Schools
For a post-deadline implementation checklist, read Ohio HB 96 School AI Policy Requirements After July 2026.
Tennessee
Classification: Binding law, local policy, annual reporting, and teacher-training duties.
Tennessee Public Chapter 550, enacted in 2024 through HB 1630 and SB 1711, requires each local board of education and public charter school governing body to adopt a policy on AI use by students, teachers, and staff for instructional and assignment purposes. The policy was to be implemented no later than the 2024-2025 school year. By each July 1, the covered governing body must report compliance to the Department of Education, including the adopted policy and how it will be enforced in the upcoming school year.
Public Chapter 1056, enacted in 2026 through SB 0677, adds professional-development duties. The Department of Education must make at least one no-cost asynchronous course on classroom AI use available to teachers in grades 6-12. Those teachers must complete at least one department-approved course by the later of August 1, 2028, or within two years after becoming licensed in Tennessee.
District implementation questions:
- Is the current policy adopted by the correct governing body and implemented in every covered school?
- Was the policy and enforcement method included in the latest July 1 report?
- Does the policy distinguish district rules from assignment-specific educator directions?
- Which grade 6-12 teachers need an approved course, and how will completion dates be tracked?
Official source: Tennessee Public Chapter 550 bill record
Official source: Tennessee Public Chapter 1056 bill record
Utah
Classification: Binding law, state model policies and local AI and classroom-technology duties.
Utah HB 273 took effect July 1, 2026. The law requires the State Board of Education to publish a model balanced classroom-technology policy by December 1, 2026 and include a model AI-use policy. Before July 1, 2027, an LEA must adopt the state model balanced-technology policy or an amended version containing an AI-use policy with a similar structure, a grade-level framework, parent transparency, and related requirements. The governing board must hold a public meeting with public comment before adoption and submit the finalized policy and meeting confirmation to the state board.
The model AI policy must address approved student-facing AI tools, educator professional judgment, independent AI grading and high-stakes determinations, student authorization for specific instructional purposes, developmental expectations, written parent notice for classroom generative-AI use, tool purpose, harmful content, biometric surveillance, psychological profiling, parent expectations, and assignment-specific guidance. LEAs must produce and adopt an AI policy based on the model, submit it, train educators and staff, ensure school compliance, review it at least every two years, and update it as needed.
The law contains more classroom-technology provisions and exceptions than this summary can reproduce. Districts should work from the enrolled act, current Utah Code, state rules, and the model policy when published.
District implementation questions:
- Who owns the public-meeting, adoption, submission, training, and two-year review calendar?
- Which student-facing AI tools are approved, for what purpose, and by which evidence?
- How will written parent notice and the metadata-dictionary link work in practice?
- How will local grading and high-stakes decision processes preserve qualified human authority?
- Which grade-level technology rules and statutory exceptions apply to each school or course?
Official source: Utah HB 273 enrolled copy
Texas
Classification: Binding law, public-sector AI governance duties that reach school districts.
Texas Senate Bill 1964 took effect September 1, 2025. School districts are local governments for these requirements. Local governments that procure, develop, deploy, or use AI must adopt the Department of Information Resources AI code of ethics. They must also adopt the state minimum standards for heightened-scrutiny AI, review deployment and use of those systems, and provide the review to the Department on request. A local government using public-facing AI or AI that controls a consequential decision must use the state’s standardized notice on related applications, websites, and public computer systems.
The law’s separate formal impact-assessment section applies to state agencies and vendors contracting with state agencies. It should not be restated as an identical school-district impact-assessment mandate.
District implementation questions:
- Which district systems are public-facing, and which could control a consequential decision about access to a government service?
- Which systems meet the narrower heightened-scrutiny definition, and which statutory exclusions apply?
- Has the district adopted the current DIR ethics code and minimum standards?
- Where is the required standardized notice displayed?
- Can the district produce its heightened-scrutiny deployment review if DIR requests it?
Official source: Texas SB 1964 enrolled bill
Official source: Texas DIR AI templates and resources
Oklahoma
Classification: Binding law, student-facing safeguards, parent choice, and local policy duty.
Oklahoma SB 1734 was approved by the governor on May 12, 2026 and took effect July 1, 2026. It applies to public school districts and regulates student-facing AI used for instructional or educational purposes. Classroom and student-facing uses must remain under educator direction, and AI output must be reviewed by an educator or other authorized employee before it is used for instruction, feedback, assessment, or a decision affecting a student.
The law says AI may not be the primary basis for grading, discipline, placement, promotion, retention, or another high-stakes educational decision. Student-facing tools must be age- and developmentally appropriate and have a defined educational purpose. Districts must take reasonable steps to minimize student data shared with AI systems and annually disclose the AI tools and vendors used, data categories, extent of sharing, and educational purpose.
A parent or legal guardian may opt a student out of participating in student-facing AI tools at any time. An opted-out student may not be academically penalized or denied access to core instructional content. Before the 2027-2028 school year, each district board must adopt a policy that identifies responsible owners, appropriate and prohibited use, data protection and minimization, transparency, periodic review, and compliance with applicable law and guidance.
District implementation questions:
- Which uses are actually student-facing, and which employee owns each approved use?
- How will the district record annual disclosure, parent choice, effective date, and alternative instructional path?
- Which workflow proves that a qualified human reviews output before feedback, assessment, or a student-affecting decision?
- How will the district prevent a parent choice from becoming a penalty or loss of core content?
- Which data fields can be removed before an approved AI system receives the request?
Official source: Oklahoma SB 1734 enrolled act
Official source: Oklahoma SB 1734 bill history
Oklahoma implementation guide: parent opt-out, alternative instruction, and board policy
North Carolina
Classification: Binding law, state model, local policy, teacher-training, and standards duties.
North Carolina Session Law 2026-41, Section 7.39, generally took effect July 1, 2026. It requires the Department of Public Instruction to develop a model policy for responsible student and staff use of AI in public schools by December 31, 2026. After reviewing that model, each public-school-unit governing body must adopt a student and staff AI-use policy by June 30, 2027. Covered units include local school administrative units, charter schools, regional schools, laboratory schools, and state-operated special schools.
The law also directs the state to develop AI professional development and requires teachers employed by specified public school units to complete it by June 30, 2028. AI literacy is scheduled to enter the K-12 computer-science standards beginning with the 2028-2029 school year.
District implementation questions:
- Who will review the state model when it is published and prepare the local adoption record?
- Which existing acceptable-use, academic-integrity, privacy, procurement, and instructional policies need to be reconciled?
- How will required teacher professional development and completion evidence be managed?
- Who owns the later AI-literacy standards implementation?
Official source: North Carolina Session Law 2026-41, Section 7.39
Official source: NCDPI generative-AI guidance
Official guidance verified in this review
Massachusetts
Classification: State education guidance, local choice.
The Massachusetts Department of Elementary and Secondary Education published K-12 AI guidance and implementation resources in 2026. The guidance states that it is intended to support districts that choose to explore or implement AI and that the Department is not recommending or requiring AI use. It covers foundations, organizing for action, equity, legal foundations, AI literacy, academic integrity, district operations, and implementation checklists.
District action: use the guidance as a local planning framework while separately identifying binding state law, regulation, contract, board policy, and federal requirements.
Official source: Massachusetts Artificial Intelligence in K12 Schools
Minnesota
Classification: State education guidance.
The Minnesota Department of Education’s 2026 guidance presents AI as a potential learning aid rather than a replacement for educators, relationships, or student thinking. It emphasizes people-centered decisions, privacy and data protection, intentional tool evaluation, local governance, professional learning, and age- and context-appropriate use.
The document is a district planning resource. It should not be recast as a statute or as proof that a particular tool is approved statewide.
District implementation questions:
- Which local instructional goals justify an AI use, and what non-AI path remains available?
- What privacy, accessibility, safety, and evidence review occurs before tool approval?
- How will local guidance preserve educator judgment and student thinking rather than automate them away?
Official source: Minnesota Department of Education AI Guidance
Oregon
Classification: State education guidance and policy-development resources.
The Oregon Department of Education maintains a current web-based Generative AI in K-12 Classrooms guide. ODE presents it as a flexible resource rather than a single required implementation path. It recommends district-level planning, local policy and protocols, cross-functional tool review, staff capacity, privacy review, human responsibility, equity, academic-integrity work, and transparent family engagement.
District action: compare existing AI rules with the current ODE guide and Oregon-specific privacy requirements, then document which recommendations the district adopts and who owns review.
Official source: Oregon Generative AI in K-12 Classrooms Guidance
Washington
Classification: State education guidance and an LEA implementation framework.
Washington OSPI publishes comprehensive Human-Centered AI Guidance for K-12 Public Schools. The current resource set includes foundations, classroom implementation, ethical considerations, definitions, FAQs, a decision rubric, and a leadership checklist. The guidance encourages an approach centered on human inquiry, responsible use, privacy, equity, safety, stakeholder engagement, and local policy development.
District action: use the OSPI framework to structure stakeholder review and local policy, then verify binding Washington statutes, rules, learning standards, privacy duties, and board policy separately.
Official source: Washington Human-Centered AI Guidance
What about the other states?
A December 2025 national review by the State Educational Technology Directors Association and Digital Promise identified official AI-in-education guidance in 32 states and Puerto Rico at that time. That broader inventory included Alabama, Alaska, Arizona, Colorado, Connecticut, Delaware, Georgia, Hawaii, Indiana, Kentucky, Louisiana, Maine, Michigan, Mississippi, Missouri, Nevada, New Jersey, New Mexico, North Dakota, Virginia, West Virginia, Wisconsin, and Wyoming in addition to states detailed above.
That report is useful for discovery, but it is not a substitute for current legal research. Guidance may be revised, withdrawn, superseded, or accompanied by later statutes. This tracker gives detailed treatment only where the current primary source was rechecked and its legal status could be stated carefully. We will expand it in reviewed batches rather than create 50 thin pages that imply certainty the sources do not support.
National review: State Guidance for Generative AI in Education
What this tracker does not establish
This page does not establish that:
- unlisted states have no AI law, rule, standard, or guidance;
- a guidance document is legally binding;
- a listed statute is the only state law relevant to an AI use;
- compliance with an AI-specific policy satisfies student-privacy, civil-rights, accessibility, records, procurement, labor, security, or local-board duties;
- a pending bill will become law;
- an approved vendor or product is approved for every purpose, user, account, data set, or model deployment.
State requirements change through legislation, rulemaking, agency interpretation, litigation, and local implementation. Preserve a dated source record and route legal conclusions to qualified counsel.
A district update workflow
- Assign an owner for state legislative, rule, standards, and agency monitoring.
- Record the exact official source, legal status, effective date, covered entity, action, and deadline.
- Map each duty to a district owner, policy, procedure, evidence record, and board calendar.
- Update the K-12 AI acceptable use policy checklist when local rules change.
- Route new duties through the district AI governance committee charter.
- Recheck product coverage in the dated Tenet supported-product capability matrix, then review contract and data implications with the AI vendor and DPA questions and data-boundary framework.
- Add new reporting, notification, and response steps to the K-12 AI incident response playbook.
- Publish a plain-language change notice for educators, students, and families when local practice changes.
Frequently asked questions
Which states require school districts to adopt an AI policy?
In this verified set, Ohio, Tennessee, Utah, North Carolina, Oklahoma, and Maryland impose local policy duties or deadlines, although their scopes and trigger dates differ. Texas imposes separate public-sector AI governance duties. California and Illinois laws direct state-level work rather than creating the same blanket local policy mandate.
Do any state school AI laws require a parent opt-out?
Oklahoma SB 1734 gives a parent or legal guardian the right to opt a student out of participating in student-facing AI tools at any time. The student may not be academically penalized or denied core instructional content, so the district needs an alternative-instruction workflow. California’s model policy recommends annual guardian consent for autonomous student-facing AI, but CDE labels the model exemplary and nonmandatory.
Does California require every district to teach a standalone AI course?
No. California AB 2876 directs the Instructional Quality Commission to consider AI literacy in specified curriculum frameworks and instructional-material criteria at future revision and adoption points. The cited law does not require each district to create a standalone AI course.
Does Texas require school districts to conduct the same AI impact assessment as state agencies?
No. Texas SB 1964 requires local governments, including school districts, to review deployment and use of heightened-scrutiny AI and provide that review to the Department of Information Resources on request. The separate statutory impact-assessment section applies to state agencies and their vendors.
Is state education-agency AI guidance legally binding?
Not automatically. Guidance can be influential and useful without carrying the same legal status as a statute or rule. Read the document’s own authority language and check state law, regulation, board policy, and local requirements.
Why are only 13 states covered in detail?
The detailed entries are limited to official sources that were rechecked and could be classified confidently as law, state duty, model policy, or guidance. The national review cited above is a discovery aid, not proof of each state’s current legal requirements.
Are pending state AI bills included as current requirements?
No. Introduced, pending, failed, and vetoed bills are not presented as current duties. They belong on a separate watchlist until an official enacted source establishes otherwise.
How should a district use this state AI policy tracker?
Use it to identify official starting sources and implementation questions, then verify the current code, rules, agency updates, local policy, effective dates, and legal interpretation with qualified district reviewers.
How often should state AI requirements be reviewed?
Review before each school year, during the legislative and rulemaking cycle, and after a material agency, court, product, privacy, or district-policy change. Record the source and review date rather than relying on an undated summary.
Official sources
- California AB 2876, Chapter 927
- California SB 1288, Chapter 893
- California State Model Policy on the Use of AI in Education
- Illinois Public Act 104-0399
- Illinois State Board of Education AI Guidance
- Maryland SB 720, Chapter 634
- Ohio Revised Code Section 3301.24
- Ohio AI Model Policy for Districts and Schools
- Tennessee HB 1630 and SB 1711, Public Chapter 550
- Tennessee SB 0677, Public Chapter 1056
- Utah HB 273 enrolled copy
- Texas SB 1964 enrolled bill
- Texas DIR AI templates and resources
- Massachusetts Artificial Intelligence in K12 Schools
- Minnesota Department of Education AI Guidance
- North Carolina DPI generative-AI guidance
- North Carolina Session Law 2026-41, Section 7.39
- Oklahoma SB 1734 enrolled act
- Oklahoma SB 1734 bill history
- Oregon Generative AI in K-12 Classrooms Guidance
- Washington Human-Centered AI Guidance for K-12 Public Schools
- State Guidance for Generative AI in Education national review
This tracker is educational information, not legal advice. Districts should verify current state code, administrative rules, education-agency updates, local policy, and applicable federal requirements with qualified reviewers.