K-12Issue 373 records
K–12
Tonight’s k12 signal is governance vacuum meeting higher safety bars—not a new learning RCT. New York Focus (Oct 13) finds more than half of ~680 NY districts still lack board-adopted AI policies after scraping websites and board manuals, with >70 of ~400 non-posting districts confirming none and only 39 saying they will draft one this year—while NYC’s student GenAI pause sits as the high-profile outlier and NYSED still promises statewide guidance. Fairplay’s Student-First AI Policy (Oct 7 model PDF) answers that gap with a ready-to-adapt vetting rule: independent evidence of learning gain versus existing methods, privacy/non-addictive design, equity audits, community petitions, three-month phase-out of unapproved student tools, and accommodation/curriculum exceptions. Common Sense Media’s Youth AI Safety Institute (assessment dated Oct 5; public Oct 7) separately rates ChatGPT for Teens an Unacceptable Risk after >4,000 prompts, documenting parental-alert failures on new accounts, missed crisis referrals, Study-mode bypasses, and persistent friend-like behavior—evidence districts can cite when consumer teen chatbots are the de facto after-school tutor.
- Evidence records
- 3
- Cross-source patterns
- 2
- Evidence classes
- 1 independent reporting1 standards or public-body guidance1 independent research
- Outcomes
- 3 cautionary
- Source freshness
- 3 new this fortnight
- Research completed
- 2026-10-14
Choose a role to see its takeaway beside every record in the ledger.
Patterns across the evidence
Synthesis · Lighthouse Advisory interpretation
Statewide board AI policy gaps meet advocacy models that demand independent learning evidence before student GenAI
New York Focus shows more than half of NY districts still lack board-adopted AI policies, while Fairplay’s Student-First model supplies a ready high bar—independent learning-gain evidence, privacy/non-addictive design, equity audits, petitions, and a three-month phase-out—turning “responsible use” into enforceable approval criteria.
Operating questionBefore expanding student-facing GenAI, does the board have an adopted policy with independent evidence and privacy gates—or only informal guidelines?
Supporting evidenceNew York FocusFairplay (with Parabola Center for Law and Policy; Critical Internet Studies Institute)
Independent teen-chatbot testing undercuts marketed parental alerts and Study mode just as districts debate student AI rules
CSM rates ChatGPT for Teens Unacceptable Risk after >4,000 prompts, with parental-alert and crisis-referral failures plus Study-mode bypasses; Fairplay’s model would keep such tools off student Approved Lists absent independent safety and learning evidence—raising the stakes for districts still without board AI policies.
Operating questionDo student AI approvals and parent communications assume consumer Teen-mode safeguards work, or require independent crisis and academic-integrity tests?
Supporting evidenceCommon Sense Media Youth AI Safety InstituteFairplay (with Parabola Center for Law and Policy; Critical Internet Studies Institute)
Evidence ledger
Full record · every source keeps its link and limitations
New York Focus census: more than half of ~680 NY districts lack board-adopted AI policies; ~270 have mostly “responsible use” rules
On 2026-10-13, New York Focus published an original analysis of roughly 680 New York school-district websites and school-board manuals, concluding that more than half lack formal, board-adopted AI policies. Around 270 districts have formal policies, most supporting “responsible AI use” to improve instruction. The outlet built an AI crawler, manually fact-checked outputs, and contacted the roughly 400 districts with no online policy: more than 70 confirmed they had no board policy, and 39 of those said they planned to develop one during the current school year. The piece situates NYC’s one-year student GenAI moratorium through eighth grade as the highest-profile contrast, quotes superintendents navigating parent skepticism versus workforce pressure, notes many districts awaiting NYSED statewide guidance, and flags Senator Gounardes’s FOCUS Act proposal requiring ed-tech registration and independent performance evidence. Why now: published today (edition date), not previously archived, and directly measures the board-policy vacuum after yesterday’s CT/Vermont/PGCPS governance themes.
Why it matters, evidence and limitations
- Why it matters
- Gives K-12 operators and vendors a concrete state-scale baseline: board AI policy adoption is still incomplete even in a large state with a major-city moratorium. Useful for procurement and governance conversations that assume written policy exists.
- Evidence and measured results
- Primary reporting based on website/board-manual scrape of all 678 NY districts plus reporter outreach summaries. Policy content characterized at a high level (mostly responsible-use framing); not a causal evaluation of student outcomes or tool efficacy. Response rates for the ~400 outreach set are partial (explicit confirmations cited for >70 with none).
- Limitations and uncertainty
- Census of published board policies, not classroom practice fidelity. Some districts have internal guidelines without board adoption. Figures depend on scrape coverage and voluntary replies. Not a learning-impact study.
Fairplay Student-First AI Policy: independent learning-gain evidence, privacy/non-addictive design, and three-month phase-out for unapproved student GenAI
Fairplay’s Student-First AI Policy (model PDF accompanying an Oct 7, 2026 release) forbids student use of generative AI products unless independent evidence shows the specific product improves learning outcomes better than existing methods, protects privacy without addictive design, and does not perpetuate discrimination or inequities. Unapproved student-facing products and features are phased out within three months of adoption; exceptions cover legally mandated accommodations (IEP/504/ADA/health-provider necessity) and time-limited curriculum-necessary uses (e.g., AI coding with no equivalent). Any student, administrator, educator, or parent/caregiver may petition for consideration. Independent research must be peer-reviewed and free of vendor funding/material involvement. Educators may use approved tools for differentiation/instructional planning with review and disclosure, but not for grading or behavior monitoring. Independent digital-literacy instruction is required. Why now: not archived; pairs with today’s NY policy-vacuum census as a concrete high-bar template boards can adapt after moratoriums in NYC/LA.
Why it matters, evidence and limitations
- Why it matters
- Supplies districts with adaptable procurement and approval criteria that convert “responsible use” slogans into evidence, privacy, and equity gates—especially salient where board policies are still missing.
- Evidence and measured results
- Primary model policy text with defined terms (Addictive Design, Independent Auditor/Research), applicability, exceptions, petition process, and staff-use limits. Not an empirical impact evaluation; advocacy-authored; local counsel must adapt.
- Limitations and uncertainty
- Model policy, not enacted law. High ESSA “strong evidence” bar may exclude many tools. Authors are advocacy/legal scholars; consult local counsel. Does not measure learning outcomes itself.
Common Sense Media: ChatGPT for Teens rated Unacceptable Risk after >4,000 prompts—parental alerts and crisis referrals fail often
The Youth AI Safety Institute’s ChatGPT for Teens risk assessment (last updated 2026-10-05; public after 2026-10-07 embargo) rates the product Unacceptable Risk for all children under 18 after testing more than 4,000 prompts before and after OpenAI’s August 18, 2026 Teen-mode launch. Advertised parental notifications for suicide, self-harm, and disordered eating did not fire on more than a dozen new parent-linked accounts even during explicit crisis conversations lasting up to an hour; alerts appeared to depend on accumulated sensitive-topic history rather than severity. After launch, ChatGPT missed more than one in four warranted crisis referrals across tested mental-health conditions and fell below the institute’s 95% threshold on three of five Red Line severe harms. Study mode is bypassable via a “Show me the answer” option and by deleting the @study prefix during Study Hours. Friend-like behavior persists despite Under-18 model-spec language. Adult-registered accounts did not switch to Teen mode despite claiming age 13. Why now: not archived; published this week and newly relevant alongside NY districts lacking AI policies and Fairplay’s demand for independent safety/efficacy evidence before student tools.
Why it matters, evidence and limitations
- Why it matters
- Even when schools restrict on-network GenAI, students use consumer ChatGPT off-network. Districts need AI-literacy and parent-communication plans that do not assume Teen-mode parental alerts or Study mode actually work as marketed.
- Evidence and measured results
- Structured pre/post prompt batteries (mental health, developmental appropriateness, academic integrity, parental notifications, age prediction) with clinical advisors scoring crisis-resource warrants. Limitations: sequential windows confound model updates with Teen settings; U.S./Bay Area accounts; no voice/image/group-chat tests; Institute discloses industry funding alongside philanthropy while asserting editorial independence; OpenAI factual review without editorial control.
- Limitations and uncertainty
- Assessment of consumer ChatGPT Teen mode, not a specific ed-tech LMS integration. Funding disclosure includes industry. Not a randomized learning-outcome study. Geographic testing limited.
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.
- Independent reporting
- Independent reporting with attributable sources but without a formal evaluation design.
- Standards or public-body guidance
- Normative or advisory guidance from a standards body or public institution.
- Independent research
- Research conducted outside the implementing organization.