K-12Issue 315 records
K–12
Tonight's K-12 signal is that student AI is off by default unless an adult allows it. Utah's draft rule would bar student generative AI for schoolwork unless an educator authorises it, require approved-tool inventories and parent notice, and tie compliance to state technology funds. In Utah, Canyons opened Gemini to high schoolers under a per-assignment stoplight while Provo keeps chatbots off student Chromebooks. Frederick County's board voted 6-1 to ban AI for students in all grades. On evidence, a new University of Maryland study links regular use of a teacher-mediated AI math platform to higher MAP scores in an Arizona district, but the design cannot rule out selection bias.
- Evidence records
- 5
- Cross-source patterns
- 2
- Evidence classes
- 3 independent reporting1 academic research1 standards or public-body guidance
- Outcomes
- 4 emerging1 cautionary
- Source freshness
- 5 new this fortnight
- Research completed
- 2026-10-08
Choose a role to see its takeaway beside every record in the ledger.
Patterns across the evidence
Synthesis · Lighthouse Advisory interpretation
Student generative AI is defaulting to off, with access only where educators authorise it
Utah's draft rule would bar student generative AI for academic work unless an educator authorises it for a specific purpose. In Utah, Canyons opens Gemini only to high schoolers and uses a per-assignment stoplight, while Provo keeps chatbots off student Chromebooks. Frederick County's board went further, banning AI for students in all grades. Settings differ, but in each case student access is off unless an adult explicitly allows it.
Operating questionWho in our district can authorise student AI use, for which grades and assignments, and how is that authorisation recorded and enforced?
Supporting evidenceUtah State Board of Education (USBE)KSL.com; sources include Canyons School District (digital teaching and learning specialist Emma Moss) and Provo City School DistrictThe Frederick News-Post; Frederick County Board of Education (Maryland)
Rules are starting to require evidence of educational value, and new district evidence shows both what is possible and its limits
Utah's draft requires that instructional technology be supported by evidence or demonstrated value, with records of effectiveness. The Goblins study shows what district data can provide: usage linked to MAP growth for 6,065 students, with larger associations at higher use. It also shows the limits: an observational design open to selection bias, 13% of users unmatched, and no separate measure of the AI tutor. Meeting evidence duties will take planned evaluation, not just usage dashboards.
Operating questionFor each student-facing AI tool, what evidence of value will we accept, and have we built the data linkage and comparison group to produce it?
Supporting evidenceUtah State Board of Education (USBE)Center for Educational Data Science and Innovation, University of Maryland, College Park (Anika Alam, Nicolas Dominguez-Mel, Jing Liu); EdWorkingPapers, Annenberg Institute at Brown University; funded by Accelerate
Evidence ledger
Full record · every source keeps its link and limitations
University of Maryland working paper: Deer Valley (Arizona) middle schoolers who used the teacher-mediated Goblins AI math platform regularly gained 0.11 SD on MAP math, but the design is observational
In Deer Valley Unified (over 32,000 students, 42 campuses), Goblins, a teacher-mediated AI math platform where students type, draw or speak and get personalised feedback, was offered to all middle-grade math teachers in 2025-26; 72.6% activated licences. Of 6,065 grade 6-8 students, 50.9% averaged at least two problems a week. Using a preregistered, machine-learning weighted propensity score design, students at that threshold scored 0.109 SD higher on within-district standardised spring MAP math (0.237 SD on NWEA's Conditional Growth Index). Students solving 10-19 and 20+ problems a week showed 0.18 SD and 0.31 SD. Students with IEPs, Hispanic students and English learners showed additional growth of 0.122, 0.084 and 0.191 SD (exploratory).
Why it matters, evidence and limitations
- Why it matters
- Answers a gap from recent editions: a new U.S. K-12 student-achievement study of an AI tool in normal district use. It complements the archived Tennessee Khanmigo RCT, where low engagement limited gains, by showing larger associations where about half of students reached a modest usage threshold, but with weaker causal design.
- Evidence and measured results
- Quasi-experimental, preregistered (OSF) weighted propensity score analysis of district administrative and platform data for one school year; baseline equivalence meets the WWC threshold according to the authors. The authors stress it is observational and vulnerable to unobserved selection; usage measures all platform problems, not AI-tutor use specifically.
- Limitations and uncertainty
- Single district; observational with possible self-selection; teachers could turn off the AI tutor (an assessment mode added mid-year), so estimates are not AI-tutor effects; 13% of users unmatched; comparison students may have used similar edtech or Goblins-made materials; subgroup results exploratory. Vendor partnership acknowledged; study funded by Accelerate.
Utah State Board of Education draft rule R277-334 (Draft 1, 5 Oct 2026): districts must keep approved AI tool inventories, bar student generative AI unless an educator authorises it, notify parents and certify compliance
Draft 1 of R277-334, Classroom Technology and Artificial Intelligence Governance, dated October 5, 2026, implements H.B. 273 (2026). LEAs must adopt the Board's model AI use policy or a compliant local policy before July 1, 2027, keep an inventory of approved student-facing AI tools, ban biometric surveillance and psychological profiling, and prohibit student generative AI for academic work unless an educator authorises it for a specific purpose. Educators may not use generative AI to independently grade work or make high-stakes determinations. Parents get written notice with a metadata-dictionary link when generative AI is used. High-risk AI interactions may not be used with students. LEAs must train staff, keep records and certify compliance to receive certain state technology funds.
Why it matters, evidence and limitations
- Why it matters
- A new state rule draft turns AI policy into auditable operating duties for every Utah LEA, as Utah districts diverge on student Gemini access. Several duties, such as tool inventories, parent notice, educator authorisation and training records, mirror requirements already in the archive for Florida, Oklahoma and Maryland.
- Evidence and measured results
- Primary-source draft rule text with tracked changes, plus the Board memo and draft model policy posted for the October 2026 committee meeting. It is a proposal at first reading; it reports no outcomes.
- Limitations and uncertainty
- Draft at first reading; text may change before final adoption. The draft defines 'high-stakes artificial intelligence interaction' but its prohibition refers to 'high-risk', a drafting inconsistency to watch. No fiscal or workload estimate in the materials opened.
KSL: With Google Gemini available to all Utah public schools, Canyons opens it to high schoolers under a stoplight protocol while Provo keeps chatbots off student Chromebooks
Utah made Google Gemini available to all public schools for 2026-27 and left implementation to districts. Canyons enabled Gemini for teachers, staff and high school students through educational accounts; younger students do not have access. A red, yellow and green stoplight protocol tells students whether AI is barred, allowed only in a teacher-defined way, or approved for each assignment. Classroom uses include AI feedback on lab analyses that students must accept or reject with reasons, and home-language overviews for a multilingual student. Canyons began in 2023 with a guidebook and educator training. Provo told parents Gemini and other chatbots are unavailable on student Chromebooks while its policy is developed.
Why it matters, evidence and limitations
- Why it matters
- Shows that a statewide platform offer produces very different student access by district, just as USBE drafts a rule requiring educator authorisation for student generative AI.
- Evidence and measured results
- News report with district statements; no usage, outcome or incident data.
- Limitations and uncertainty
- Two districts described; claims are district statements; no data on compliance with the stoplight protocol or learning effects.
Frederick News-Post: Frederick County Public Schools board votes 6-1 to ban AI for students in all grades after teachers urge a full ban
At its Wednesday, Oct. 7 meeting, after hearing from some Frederick County Public Schools teachers, the Frederick County school board voted 6-1 to ban AI for students in all grades; Vice President Rae Gallagher voted against. A photo caption shows an Urbana High School teacher advocating a full student ban during the meeting. The archive shows the same board pressing in September to disable Google Gemini on student devices while it wrote its policy under Maryland's AI law.
Why it matters, evidence and limitations
- Why it matters
- A Maryland district board adopted an all-grades student AI ban by formal vote while writing the policy Maryland law requires. It contrasts with Utah districts opening Gemini to high schoolers and with state rules that allow educator-authorised use.
- Evidence and measured results
- Same-day local news report. Only the headline, photo caption and first two paragraphs were readable without a subscription; the rest of the article was not accessed, so exceptions, enforcement and timing are unknown from this source.
- Limitations and uncertainty
- Partial access: policy text, exceptions, effective date and implementation steps were not inspected. Single district; news report of a vote, not outcomes.
WNDU: South Bend Community School Corporation survey finds 74% of grade 6-12 students use AI and 70% of teachers worry about ethical use, as the district acquires MagicSchoolAI with a state grant
At its Monday board meeting, South Bend reported a district survey: 74% of almost 1,200 students in grades 6-12 said they use AI, mostly to find answers or explain what they are learning; 70% of 147 teachers were concerned about ethical and correct use, and nearly a third used AI for lesson planning. Over half of 133 responses rated existing teacher and principal AI guidelines adequate on access, integrity, implementation and data security. After testing MagicSchoolAI and SchoolAI last year, with MagicSchoolAI used most, the district won a digital learning grant to acquire MagicSchoolAI for 2026-27.
Why it matters, evidence and limitations
- Why it matters
- Gives fresh U.S. district data on how widely secondary students already use AI, and closes a gap from the 2026-10-06 run, when the only account found (South Bend Tribune) was paywalled.
- Evidence and measured results
- News report of a district survey presented to the board; survey method, timing and response rates not reported; the 133-response figure appears to be a separate guidelines question.
- Limitations and uncertainty
- Published late on Oct. 5 (local time), before the last run, and included because last night's only source was paywalled. Survey methods unknown; self-reported use.
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.
- Academic research
- Research produced through an academic institution or peer-reviewed venue.
- Standards or public-body guidance
- Normative or advisory guidance from a standards body or public institution.