{"resourceId":"instructure-csu-systemwide-genai-rollout-lessons-20261006","versions":[{"version":"external-664425108929b7e799fe9df903aa67a37cb35d991b0f74a6b25a8e8216785142","resource":{"id":"instructure-csu-systemwide-genai-rollout-lessons-20261006","title":"CSU's system-wide AI rollout: access reached 22 campuses, but students still ask whether AI is allowed and fear tracking","organization":"California State University system (as told to Instructure; vendor content-marketing blog)","sector":"Higher education student success / system-wide GenAI access, AI literacy and policy clarity","geography":"United States (California, 22-campus public university system)","publishedAt":"Published 6 October 2026","publicationDate":"2026-10-06","eventDate":null,"sourceName":"Instructure blog","sourceLabel":"Vendor content-marketing blog (Instructure, CSU's LMS provider) interviewing a CSU academic technology leader","sourceUrl":"https://www.instructure.com/resources/blog/safety-responsibility-and-curiosity-how-put-ai-front-half-million-students","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["governance-procurement","data-security","accessibility-workforce","operating-model"],"finding":"CSU's assistant vice chancellor for academic technology services describes a system-wide initiative that began with an advisory committee of faculty, students and staff and an equity goal: the system bought an enterprise AI environment for everyone, with privacy and IP protections and no user data feeding a model, reaching all 22 campuses in 18 months. 'Access, it turned out, was the easy half'; the work shifted to AI literacy trainings in the central Canvas environment, an AI Commons, funded faculty AI fellows and town halls. Campuses built custom GPTs for tutoring, applicants and advising aimed at retention and are now experimenting with agentic models. The system's survey last fall found use alongside concerns about environmental cost and privacy; some students resist the system's OpenAI instance because they believe they'll be tracked, and students say the biggest question is whether AI is allowed in a given class. The systemwide senate issued a recommendation for faculty; only some campuses publish a syllabus policy. No outcome data are reported.","sledRelevance":"Published today, it is the largest U.S. public-system account of what follows equitable AI access: students still distrust the sanctioned tool on privacy and still don't know course-level rules. It complements the Alaska case, where rules are also course-by-course, and should be read as vendor-told experience, not evaluation.","evidence":"Vendor blog summarising an interview (and podcast episode) with a CSU administrator; references the CSU system survey without figures. The leader's surname appears as both 'Kennedy' and 'Kenney' on the page. No measured retention, learning or usage outcomes.","architectureImplications":"A single enterprise AI environment with no-training terms provided system-wide access; training and resources run through the central LMS. Agentic experiments raise new integration and permission questions the post does not address.","governanceImplications":"Interpretation: access without course-level rules leaves students guessing; a senate recommendation without consistent campus syllabus policies leaves gaps. Shared-governance bodies were central to the rollout.","securityPrivacyImplications":"Despite contractual protections, some students believe they are tracked in the sanctioned instance, so trust requires explaining data handling in plain terms, not only contracting for it.","caveats":"Vendor marketing content about a customer; no outcome, usage or survey figures; single interviewee; inconsistent spelling of the interviewee's name; claims about privacy protections are not independently verified.","streamIds":["student-success"],"roles":{"sales":"The framing point is that access was the easy half. CSU's account, told on a vendor blog, says the system bought AI for every student across 22 campuses, yet students still distrust the tool on privacy and don't know course rules. Use it to scope what comes after licensing: literacy, faculty support, policy clarity and plain-language data assurances. Treat it as one leader's experience with no outcome data, and don't cite it as evidence of retention or learning gains.","engineering":"For engineering, the framing point is trust in the sanctioned environment. Contract terms excluded training on user data, yet students still believe they are tracked. Document what is logged, who can see prompts, and retention periods, and publish that plainly to students. As campuses move from custom GPTs toward agentic tools, define which systems an agent may read or act on, and route training and resources through the LMS students already use.","delivery":"Delivery's framing point is the work after rollout. Plan AI literacy training, funded faculty champions, town halls and a course-level syllabus template so every class states its AI rules. Survey students on clarity and trust each year and publish results by campus, as CSU did. Track whether students using the sanctioned tool rises after data-handling explanations, and treat custom retention-focused bots as pilots that need their own evaluation."},"roleDetails":{"sales":{"customerProblem":"After licensing AI for all students, systems still face low trust in the sanctioned tool and confusion about course-level rules.","stakeholders":["System CIO and academic technology leadership","Systemwide academic senate","Campus provosts","Student government","Procurement and privacy office"],"discoveryQuestions":["What share of students use the sanctioned AI tool versus consumer tools?","Does every course state its AI rules?","How do you explain data handling in the sanctioned tool to students?"],"valueHypothesis":"Plain-language data assurances, course-level rule templates and literacy programs may raise use of sanctioned tools and reduce confusion; effects need measurement.","potentialEngagement":"A post-rollout adoption review: trust and clarity survey, syllabus-policy coverage audit and literacy program design.","unsupportedClaims":["That system-wide AI access improves retention","That custom tutoring or advising GPTs improved outcomes","That privacy protections are verified"]},"engineering":{"fit":"Multi-campus systems with an enterprise AI licence and a central LMS.","architectureIntegration":"Enterprise AI environment under no-training terms, SSO, LMS-hosted training, and governance for campus-built custom assistants and agents.","prerequisites":["Enterprise data-processing terms","Clear logging and retention settings","Inventory of campus-built assistants"],"constraints":["Student distrust of tracking","Decentralised campus experimentation","Agentic tools with broader system access"],"security":"Publish logging, access and retention settings; restrict agent permissions; review campus-built assistants before student release.","proofOfValue":["Sanctioned-tool adoption before and after data-handling explanations","Share of courses with a published AI rule","Inventory and review status of custom assistants"]},"delivery":{"work":"Follow licensing with literacy training, faculty fellows, course rule templates and annual trust and clarity surveys.","dependencies":["Senate endorsement of templates","Funding for faculty fellows","Survey and IR capacity"],"ownership":"System academic technology office with campus teaching centers; senate owns policy guidance.","skillsAdoption":"Self-paced and facilitated literacy courses for students, faculty, staff and administrators.","governanceCheckpoints":["Senate review of syllabus template","Privacy review of logging settings","Annual survey review by campus"],"acceptanceCriteria":["Published AI rule in a target share of courses","Annual student trust and clarity survey completed and published","Every campus-built student assistant reviewed before launch"],"risks":["Access mistaken for adoption","Persistent privacy distrust","Unevaluated retention bots"]}},"retrievedAt":"2026-10-07T03:13:13Z","enrichedAt":"2026-10-07T03:17:13Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Equitable access was the stated first goal; faculty and administrators both need time, support and a safe space to learn, which the system addressed with funded fellows and training.","procurementImplications":"System-level purchasing can close access gaps, but contracts need clear no-training and data-handling terms that are communicated to students.","operatingModelImplications":"Budget for literacy, faculty fellows and course-level policy templates after licensing; access alone did not resolve student uncertainty.","sourceVerification":{"openedUrl":"https://www.instructure.com/resources/blog/safety-responsibility-and-curiosity-how-put-ai-front-half-million-students","referenceExcerpt":"Some students have pushed back on using the system's OpenAI instance because they believe they'll be tracked in it.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}