K-12Issue 525 records
K–12
Loudoun County will start a student pilot of Gemini Notebook at eight schools in November, pitching it as a study tool grounded in class materials, not a general chatbot. Valparaiso, Indiana tabled a teacher AI tool contract over how the vendor could use district data. Scarsdale is teaching AI literacy instead of adopting commercial AI tools. Providence says its AI reading tutor Amira is also its state-required dyslexia screener, which helps explain why families can't easily opt out. New York enacted limits on companion chatbot features for minors, backed by the state teachers union.
- Evidence records
- 5
- Cross-source patterns
- 2
- Evidence classes
- 3 standards or public-body guidance2 independent reporting
- Outcomes
- 3 emerging1 cautionary1 mixed
- Source freshness
- 5 new this fortnight
- Research completed
- 2026-10-29
Choose a role to see its takeaway beside every record in the ledger.
Patterns across the evidence
Synthesis · Lighthouse Advisory interpretation
Vendor data terms are deciding whether school AI tools go ahead
Valparaiso's board tabled a Playlab teacher-tool contract over how the vendor could use district data; Loudoun justified its student Gemini Notebook pilot by citing Google's no-training commitment and district accounts; Providence defended Amira by citing a no-retention setting and vendor data agreements, while parents still ask about earlier recordings.
Operating questionFor each AI tool, can the district state in writing what data the vendor receives, keeps, deletes and may use for training, including data collected before current settings?
Supporting evidenceValparaiso Community Schools (Indiana), as reported by Our School BoardLoudoun County Public Schools (Virginia)Providence Public School District (Rhode Island)
Districts and states are narrowing what kind of AI students may use rather than choosing all or nothing
Loudoun allows a source-grounded study tool rather than a general chatbot; Providence stresses that Amira cannot hold open conversations; Scarsdale teaches AI literacy without commercial student tools; New York bans specific companion features for minors.
Operating questionWhich specific AI behaviors (open conversation, persona, memory, grounding) are allowed for which grades, and how is each verified in the tools students can reach?
Supporting evidenceLoudoun County Public Schools (Virginia)Providence Public School District (Rhode Island)Scarsdale Union Free School District (New York), as reported by Our School BoardOffice of the Governor of New York; NY State Legislature
Evidence ledger
Full record · every source keeps its link and limitations
Loudoun County Public Schools announces November student pilot of Gemini Notebook at eight schools, framed as a source-grounded study tool under Workspace for Education terms
LCPS announced that eight schools will begin piloting Gemini Notebook (formerly NotebookLM) with students in November. The district says the tool grounds answers in teacher-provided or student-uploaded materials, unlike general-purpose chatbots, and will be used for study guides, practice quizzes, audio overviews and outlining. Access runs through student LCPS accounts in the district's Google Workspace for Education environment; the release cites Google's commitment that student content is not used to train public models. Teachers will guide responsible use, and the district will gather teacher and student feedback. The release does not name the schools, grades, duration or success measures.
Why it matters, evidence and limitations
- Why it matters
- Directly K-12 and current: it moves the Loudoun pilot archived on Sept 22 (optional secondary pilots amid a split board) to a dated launch with stated privacy terms and use cases. It is a counterpoint to nearby Frederick County, which banned student AI after a Gemini pilot.
- Evidence and measured results
- Official district announcement; describes intended uses, access model and feedback collection. No evaluation design, metrics, sample or results.
- Limitations and uncertainty
- District self-description; grounding reduces but does not remove inaccurate outputs, as the release itself tells students to verify. 'Feedback' is not an outcome measure. Schools and grades not named.
Valparaiso (IN) school board tables Playlab AI teacher tool contract over district data use and legal terms; district still has no AI rules
On Oct. 15 the Valparaiso Community Schools board tabled a contract with Playlab, described in the meeting as the district's AI software and training tool for teachers, which would have run through June 30. Board members raised concerns about how Playlab could use district data and about the contract's legal terms. The summary says the district has no AI rules yet, staff said guidelines had been in development for about 18 months, and a parent in June called high school messaging about AI punitive and negative.
Why it matters, evidence and limitations
- Why it matters
- K-12 and current: a board is pausing a teacher AI tool at the contract stage over data-use terms, showing procurement review as the practical gate for AI adoption in districts without an adopted AI policy.
- Evidence and measured results
- Summary of a board meeting recording based on auto-generated captions; no contract text, cost or follow-up action was available.
- Limitations and uncertainty
- The outlet states it uses AI to draft summaries ('AI can make mistakes') from auto-captioned recordings; contract value and the vendor's response are unknown. Published shortly before the last run but not archived; included because procurement pauses are a current trend.
Scarsdale (NY) pilots AI-literacy lessons in grades 6-8 and elementary library projects instead of adopting commercial classroom AI tools
At the Oct. 26 board meeting, Scarsdale staff described an initial AI-literacy unit piloted in grades 6 and 7 at Scarsdale Middle School, with grade 8 choosing to include the lessons, and elementary AI-literacy pilots embedded in library and technology research projects. Lessons cover AI-generated misinformation, bias, responsible use and when not to rely on AI. Staff said the district was not rushing to adopt commercial AI products.
Why it matters, evidence and limitations
- Why it matters
- K-12 and current: a district is choosing to teach about AI before deploying AI tools to students, a distinct path from both tool pilots and bans.
- Evidence and measured results
- Summary of a board meeting transcript (cited at 1:57-1:59); no curriculum materials, hours or assessment of learning were available.
- Limitations and uncertainty
- Automated summary ('AI can make mistakes') of a short segment of a meeting; curriculum content, duration and evaluation are not described. Single affluent district.
Providence Public Schools statement: Amira is the district's state-required dyslexia screener, new voice recordings are not retained, and platforms take about 7% of weekly instructional minutes
In a statement to families, PPSD says K-5 students use Amira, which listens as students read aloud and gives feedback on accuracy and fluency, and cannot hold everyday conversations. The district recommends 30 minutes of weekly practice, has selected a setting under which new student voice recordings are not retained, and uses Amira as its dyslexia screening tool, which the Rhode Island Department of Education (RIDE) requires. It says 110 minutes of weekly platform use is about 7% of 1,650 instructional minutes, and that it protects data through vendor data sharing agreements, privacy partnerships and application-blocking tools.
Why it matters, evidence and limitations
- Why it matters
- Newly relevant: the statement was posted before the last run, but parents' Oct. 21 board testimony (archived Oct. 26) and Oct. 28 TV coverage keep the dispute live. It adds the district's side, notably that Amira doubles as a state-required dyslexia screener, which explains why opting out is hard.
- Evidence and measured results
- Official district communication with stated settings and time-use figures; it does not address recordings made before the August setting change, consent history or effectiveness evidence.
- Limitations and uncertainty
- District's own account; dated Oct. 22, before the last run, and included because the controversy continued through Oct. 28. Does not answer parents' questions about pre-August data. Effectiveness of Amira is not addressed.
New York enacts S.9051B/A.10379C barring unsafe AI companion chatbot features for minors, with age verification and attorney-general enforcement; NYSUT backed the bill
Gov. Hochul signed S.9051B/A.10379C, which prohibits companies offering AI chatbot features from giving minors, or any user not confirmed as an adult, features such as pretending to be human, claiming personal, romantic or authority relationships, unprompted emotional questions, reuse of a child's private health or personal information from earlier chats, promoting self-harm, encouraging secrecy or isolation, and sexual content. Companies must offer more than one age verification method, including one not solely relying on government ID, and delete age data after use. The attorney general can seek penalties up to $25,000 per violation. NYSUT's president praised the law.
Why it matters, evidence and limitations
- Why it matters
- K-12 relevant for school-age learner safety: most teens use AI companions according to cited Common Sense Media data, and the law sets feature-level limits that districts can reference when evaluating student-facing chatbots. It also shows teacher unions lobbying on youth AI safety beyond classrooms.
- Evidence and measured results
- Official press release summarizing bill provisions and quoting sponsors, the attorney general, Common Sense Media and NYSUT; the bill text and effective date for the chatbot provisions were not inspected.
- Limitations and uncertainty
- Press release, not statute text; the release says 'in the coming months' and gives no exact effective date for the chatbot provisions. Applicability to school-procured tools is unclear. Teen usage figures are attributed to Common Sense Media and were not verified here.
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.