{"resourceId":"digitalpromise-k12-ai-infra-cohort2-20260921","versions":[{"version":"external-689f1604ddb0922a7b741bc83b2108bcc19e536287cd23ace3141747a375129b","resource":{"id":"digitalpromise-k12-ai-infra-cohort2-20260921","title":"Digital Promise: second K-12 AI Infrastructure cohort funds open formative-assessment datasets and privacy-preserving writing feedback","organization":"Digital Promise K-12 AI Infrastructure Program","sector":"Nonprofit education research / public-goods AI infrastructure funding","geography":"United States (multi-institution grantees including rural National Trail Local SD, OH)","publishedAt":"21 September 2026","publicationDate":"2026-09-21","eventDate":null,"sourceName":"Digital Promise","sourceLabel":"Program announcement of second grant cohort (HTML)","sourceUrl":"https://digitalpromise.org/2026/09/21/new-grantees-awarded-by-the-k-12-ai-infrastructure-program/","evidenceClass":"independent-research","outcomeClass":"emerging","topics":["infrastructure","data-security","knowledge-work","accessibility-workforce"],"finding":"On 21 Sep 2026, Digital Promise announced eight second-cohort grantees in its $26M multi-year K-12 AI Infrastructure Program, prioritizing formative-assessment tools and deepening math/literacy/writing investment. Outputs will be openly licensed. Notable projects include National Trail Local SD’s privacy-preserving, discipline-specific AI writing feedback as a public good for under-resourced rural districts; Quill.org’s “Productive Coaching” benchmark validating coaching quality in student-facing AI; Harvard OpenLiteracy speech models for early word reading; multimodal classroom datasets (UMD, Pitt, MMSA/TERC); UF dialect-inclusive early-reader corpus; Playlab science CER writing feedback dataset. Next RFP later fall. Why now: after 09-25 Instruction Partners finding that purpose-built tools need evaluation capacity, this cohort funds shared benchmarks/datasets—including rural privacy-preserving writing—rather than closed vendor features.","sledRelevance":"Points SLED buyers toward open formative-assessment infrastructure and rural privacy-preserving writing feedback as alternatives to opaque chatbot tutoring claims.","evidence":"Funder program announcement listing projects; not completed evaluations of classroom learning outcomes.","architectureImplications":"Prefer open datasets/benchmarks and privacy-preserving feedback architectures over closed general chat; invest in multimodal formative assessment pipelines.","governanceImplications":"Public-goods licensing and competitive review with educators/researchers; track rural district as first-class builder, not only consumer.","securityPrivacyImplications":"Explicit privacy-preserving writing-feedback project signals privacy as a design requirement for student writing AI.","caveats":"Announcement stage; 6–12 month projects; no efficacy results yet; Digital Promise is the program operator.","streamIds":["k12"],"roles":{"sales":"Sales framing: after chatbot-risk coverage, sell open formative-assessment infrastructure and privacy-preserving writing feedback—not another generic tutor. Digital Promise’s 21 September second cohort funds shared datasets, Quill productive-coaching benchmarks, and a rural district public-good writing tool under open licenses. Engage on evaluation capacity and reusable outputs with CAOs and assessment leads. Do not claim these grants already prove learning gains, that open license equals classroom readiness, or that program dollars replace local professional development.","engineering":"Engineering framing: fit is multimodal formative-assessment stacks and privacy-preserving feedback services with open licensing rather than closed general chat. Prerequisites include data-governance plans for classroom audio or writing corpora, privacy threat models, and benchmark harnesses such as productive coaching validation. Constraints: projects run only six to twelve months and remain early. Proof of value is usable open artifacts plus documented privacy designs—not vendor seat growth or engagement dashboards.","delivery":"Delivery framing: track second-cohort open releases, pilot privacy-preserving writing-feedback patterns where ready, and refuse to treat grant announcements as classroom-ready products. Ownership sits with assessment/research for instructional use and IT for data pipelines and privacy threat models. Skills adoption means teachers can explain AI formative insights without outsourcing judgment. Acceptance criteria include posted open licenses, signed privacy design reviews, and defined classroom pilot gates. Risks include announcement hype, dataset bias, and rural capacity gaps that leave tools unused."},"roleDetails":{"sales":{"customerProblem":"Districts lack shared, trustworthy formative-assessment AI building blocks and fear opaque student writing tutors.","stakeholders":["CAO","Assessment directors","Rural superintendents","CTOs","Research partners"],"discoveryQuestions":["Do you need open benchmarks or another closed tutor?","Can writing feedback be privacy-preserving?","Who validates coaching quality in student AI?"],"valueHypothesis":"Open formative-assessment infrastructure reduces vendor lock-in and supports purpose-built evaluation.","potentialEngagement":"Landscape brief mapping cohort outputs to district assessment stack and privacy requirements.","unsupportedClaims":["That any grantee tool already raises scores","That open license equals classroom readiness","That $26M program replaces local PD"]},"engineering":{"fit":"Open datasets, speech/literacy models, and privacy-preserving writing feedback services.","architectureIntegration":"Connect classroom capture pipelines to formative-assessment models with strict retention controls.","prerequisites":["Data-sharing agreements","Privacy threat model","Benchmark evaluation harness"],"constraints":["6–12 month build timelines","Multimodal data sensitivity","Rural bandwidth"],"security":"Prioritize privacy-preserving designs for student writing; minimize identifiable retention.","proofOfValue":["Open artifact published","Privacy review signed","Benchmark metrics documented"]},"delivery":{"work":"Track grantee releases, pilot applicable open tools, and align with local formative-assessment practice.","dependencies":["Grantee timelines","District privacy office","Teacher PD"],"ownership":"Assessment/research owns instructional use; IT owns data pipelines.","skillsAdoption":"Teachers need formative-assessment literacy to use AI insights without outsourcing judgment.","governanceCheckpoints":["Privacy design review","Open-license verification","Pilot ethics check"],"acceptanceCriteria":["No silent PII training","Teacher can explain AI feedback basis","Open outputs reused under license"],"risks":["Announcement hype","Dataset bias","Rural capacity gaps"]}},"retrievedAt":"2026-09-27T03:09:40Z","enrichedAt":"2026-09-27T03:13:51Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Rural district as grantee builder and dialect-inclusive speech datasets aim to widen who AI literacy tools serve.","procurementImplications":"Favor open-licensed formative-assessment components and require privacy-preserving designs in writing-feedback RFPs.","operatingModelImplications":"Build evaluation capacity around shared benchmarks rather than sole-sourcing closed student chatbots.","updateExplanation":"Not previously archived. Newly relevant after 2026-09-25 archived Instruction Partners/EdWeek call for purpose-built tools under teacher orchestration: this 21 Sep cohort funds open formative-assessment datasets and a rural privacy-preserving writing-feedback public good that operationalize that direction.","sourceVerification":{"openedUrl":"https://digitalpromise.org/2026/09/21/new-grantees-awarded-by-the-k-12-ai-infrastructure-program/","referenceExcerpt":"funding returning applicants dedicated to math, literacy, and writing.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}