K-12Issue 235 records
K–12
Fresh K-12 signal on 28 Sep centers on state AI-policy mandates becoming classroom operating models—and on independent evidence that assisted output is not retained learning. WOSU’s same-day reporting shows Ohio’s first school year under ORC 3301.24 producing uneven practice: Hilliard’s AI Readiness Framework spans kindergarten through high school with assignment-level AI-use tiers (including draft-then-AI-feedback), while Columbus uses AI mainly for accuracy critique without new supplemental programs. ODEW’s model policy supplies the compliance floor—literacy, special-education-inclusive workgroups, FERPA-gated procurement, and integrity rules that AI must not replace student work—while reminding districts AI is not a substitute for student effort or educators. AEFP’s June 2026 Live Handbook synthesis (Barnard/Fesler/Loeb; newly archived here) cautions that causal U.S. student-facing evidence remains thin and that in-tool gains often fail independent proficiency tests when tools answer rather than coach. Virginia’s 2026 AIS statute raises control expectations: division-managed platforms, bans on external training with student data, teacher transcripts plus admin monitoring, and pilots evaluated for learning, teacher workload, and equity. Oregon’s GenAI Guidance v3 pairs productive-struggle preservation with explicit warnings about cognitive offloading and students forming human-like emotional attachments to AI. Net for SLED: treat policy adoption as the start of operating-model work; demand unaided mastery and anti-companion defaults; and build monitorable, division-controlled architectures before scaling student-facing generative AI.
- Evidence records
- 5
- Cross-source patterns
- 2
- Evidence classes
- 3 standards or public-body guidance1 independent reporting1 academic research
- Outcomes
- 3 emerging2 cautionary
- Source freshness
- 4 older, newly relevant1 new this fortnight
- Research completed
- 2026-09-29
Choose a role to see its takeaway beside every record in the ledger.
Patterns across the evidence
Synthesis · Lighthouse Advisory interpretation
State AI policy mandates create adoption floors; classroom depth still varies widely
Ohio’s ODEW model and same-day WOSU reporting show a statutory 'have a policy' duty producing uneven local practice—from Hilliard’s K–12 readiness continuum to Columbus’s lighter compare-and-contrast pedagogy—while Virginia’s statute raises the bar further with division-managed platforms, transcript monitoring, and evaluated pilots. Mandates alone do not equal instructional redesign.
Operating questionBeyond board adoption, can the district show assignment-level AI-use tiers, teacher-visible transcripts/monitoring where student AI is live, and a near-term evaluation plan for learning and workload?
Supporting evidenceWOSU Public MediaOhio Department of Education and WorkforceCommonwealth of Virginia (Code of Virginia)
Assisted performance is not retained learning—and companion-like AI adds developmental risk
AEFP’s synthesis finds AI often boosts in-the-moment performance without independent proficiency, especially when tools give answers rather than guide reasoning. Oregon’s guidance pairs that learning risk with warnings about cognitive offloading, lost productive struggle, and students forming human-like emotional attachments to AI. Together they demand unaided mastery checks and companion-mode limits.
Operating questionFor every student-facing AI feature, can learners demonstrate the skill without the tool—and is companion/anthropomorphic interaction disabled or separately reviewed for developmental risk?
Supporting evidenceAssociation for Education Finance and Policy (AEFP) Live HandbookOregon Department of Education
Evidence ledger
Full record · every source keeps its link and limitations
WOSU: Central Ohio districts operationalize Ohio’s first school-year AI policy mandate with literacy frameworks and use tiers
Izza Haq reports that in Ohio’s first school year under a statewide AI-policy requirement, districts are implementing unevenly: Hilliard City Schools launched an AI Readiness Framework introducing AI concepts from kindergarten through high school and teacher guidelines that define AI-use levels for assignments (including a 'full AI' path where students draft on paper then use AI for feedback before revision), while Columbus City Schools has not created supplemental AI programs but has teachers ask students to compare their work with AI to surface inaccuracy. Officials stress embedding AI literacy in curriculum, gathering stakeholder feedback, and staying agile as the technology changes; Olentangy, Westerville, and Dublin officials were unavailable for comment.
Why it matters, evidence and limitations
- Why it matters
- Shows how a state 'adopt a policy' mandate translates into concrete district operating models—useful for Ohio LEAs and other states watching statutory AI-policy requirements.
- Evidence and measured results
- Local public-radio reporting with on-record CTO/CIO quotes dated the research day; not an outcome evaluation.
- Limitations and uncertainty
- Journalistic snapshot of two districts; no learning-outcome measures; several districts did not comment; Ohio-specific statute.
Ohio ODEW model AI policy: literacy, workgroup, FERPA procurement, and integrity floor under ORC 3301.24
ODEW’s model policy, developed to help districts meet ORC 3301.24’s July 1, 2026 adoption deadline, frames AI as support for learning/teaching—not a substitute for student effort or educators—and requires AI literacy, a standing multi-stakeholder AI workgroup (including special education), FERPA/PII-aligned data controls, procurement/evaluation of AI tools for privacy/accessibility/goals/cost, context-specific acceptable use stated in syllabi with district-provided tools only, and academic integrity rules that AI-generated work must not replace student work. Related considerations flag NCII/deepfake policy integration and crosswalks to special education and other local policies.
Why it matters, evidence and limitations
- Why it matters
- Primary SEA template behind Ohio’s first-in-nation statutory AI-policy duty; pairs with same-day district implementation reporting.
- Evidence and measured results
- Official state model policy PDF retrieved in full; advisory template districts may adopt verbatim or customize.
- Limitations and uncertainty
- Model is a compliance floor, not outcome evidence; silent on many liability edge cases (e.g., detector discipline, FAPE masking); publication banner reads 2025 while deadline is 2026.
AEFP Live Handbook: K–12 AI evidence thin; assisted gains often fail independent proficiency tests
Cristina Barnard, Lily Fesler, and Susanna Loeb’s June 2026 AEFP Live Handbook chapter synthesizes that schools adopt AI faster than researchers can evaluate it: a recent review found only about 20 high-quality causal studies amid 800+ papers, with especially thin U.S. K–12 student-facing causal evidence. Key patterns include assisted performance that does not consistently become independent proficiency; design choices (guided reasoning vs answer generation) shaping learning; and educator-facing time savings whose value depends on redirecting attention to high-value human interactions. They cite RAND 2025 survey rates (~54% of secondary students and ~53% of core teachers using AI) to underscore the adoption–evidence gap.
Why it matters, evidence and limitations
- Why it matters
- Independent academic frame for distinguishing retained learning from assisted output—central to the K12 stream brief—and for challenging sales claims built on in-tool metrics.
- Evidence and measured results
- Peer-oriented research handbook chapter with cited RCTs/reviews; synthesis rather than a new primary trial.
- Limitations and uncertainty
- June 2026 update—older than last k12 run but not previously archived; why-now is direct contradiction of access-as-learning narratives after recent tutor engagement failures; many cited studies outside U.S. K–12.
Virginia Code §22.1-20.2:1: division-managed AIS, transcript monitoring, training bans, and evaluated pilots
Virginia’s 2026 statute directs VDOE to compile instructional AIS uses and post guidance covering FERPA/COPPA privacy, teacher training, transparency, DPAs that prohibit using division or student data to train external AIS models, bias/accessibility practices, and SOL/HQIM alignment. Guidance must favor division-managed AIS platforms in division-managed environments and ensure teachers can access transcripts of student–AIS interactions with administrator dashboards, automated alerts, and audit logs. School boards must implement consistent policies. An AIS Innovation in Education Pilot Program (expires July 1, 2030) funds/evaluates instruction, tutoring, engagement, operations, and teacher-support uses, prioritizing diverse/high-poverty/rural/under-resourced divisions, requiring PD plus evaluation plans for student learning, teacher workload, and equity, with annual legislative reporting.
Why it matters, evidence and limitations
- Why it matters
- Concrete statutory control stack—managed platforms, no external training, transcript supervision, workload/equity evaluation—that other states and vendors will be measured against.
- Evidence and measured results
- Primary statutory text from LIS; implementation details depend on VDOE guidance and appropriations.
- Limitations and uncertainty
- Guidance content and pilot funding not fully specified in the statute section alone; effective dates/appropriations need local confirmation; not outcome evidence.
Oregon ODE GenAI Guidance v3: human-centered use, productive struggle, and risks of AI relationships
Oregon ODE’s June 2026 Generative AI in K-12 Classrooms Guidance (v3) centers human relationships and educator agency: GenAI should strengthen, not replace, authentic classroom relationships; humans stay in the driver’s seat; and leaders should create policy guardrails before GenAI enters classrooms. Dedicated sections address students depending too much on GenAI (cognitive offloading; protecting productive struggle; AI-resistant/AI-integrated assessment redesign) and concerns about forming human-like relationships with AI, citing continuum research from helpful use to emotional attachment reinforced by warmth/validation features, with younger users potentially more vulnerable. Privacy guidance stresses never entering PII into genAI systems and FERPA/COPPA competence.
Why it matters, evidence and limitations
- Why it matters
- Directly operationalizes the K12 stream’s developmental-risk and assisted-vs-retained-learning brief with SEA language usable beyond Oregon.
- Evidence and measured results
- Official 42-page SEA guidance PDF inspected in full; synthesizes external research (Brookings, UTS, RCTs) rather than running a new Oregon trial.
- Limitations and uncertainty
- June dated—older than last run but not previously archived; advisory not statutory; cited attachment/offloading studies vary in strength; why-now is alignment with same-day state-mandate operationalization and recent mastery themes.
How to read this edition
Source findings, measured results and limitations come from the cited publications. Patterns, operating questions, role takeaways and implementation considerations are Lighthouse Advisory interpretation, stated as questions to validate locally rather than guaranteed outcomes. Vendor and operator claims are labeled as claims. Full research method.
- Standards or public-body guidance
- Normative or advisory guidance from a standards body or public institution.
- Independent reporting
- Independent reporting with attributable sources but without a formal evaluation design.
- Academic research
- Research produced through an academic institution or peer-reviewed venue.