{"resourceId":"cal-poly-ai-factory-early-adopters-protected-data-local-training-20261028","versions":[{"version":"external-3b71819e9fb9b5041a701366fc1201336b5bfef9e610f75dfb3c6a61aa6d6215","resource":{"id":"cal-poly-ai-factory-early-adopters-protected-data-local-training-20261028","title":"Cal Poly's first AI Factory users report local training on protected medical data and about 10x faster eelgrass image analysis","organization":"Cal Poly (College of Engineering, Noyce School of Applied Computing)","sector":"Higher education (teaching and research computing)","geography":"California, United States","publishedAt":"2026-10-28","publicationDate":"2026-10-28","eventDate":"2026-10-22","sourceName":"Cal Poly College of Engineering","sourceLabel":"Cal Poly Unveils NVIDIA-Powered AI Factory to Expand AI Education and Research","sourceUrl":"https://ceng.calpoly.edu/news/cal-poly-unveils-nvidia-powered-ai-factory-expand-ai-education-and-research","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["infrastructure","data-security","accessibility-workforce"],"finding":"Cal Poly's Oct 28 article on its $3M AI Factory (four NVIDIA DGX B200 systems, unveiled Oct 22) reports first faculty uses: a drone-imagery eelgrass system that matched hand annotation quality while processing about 10 times faster, and spinal cord injury and genetic risk projects that train locally because protected patient information cannot be sent to outside AI services. Faculty frame owning hardware as control over teaching and research.","sledRelevance":"Adds early-adopter evidence to the previously archived Cal Poly AI Factory announcement: the clearest SLED-relevant driver is protected data that cannot go to commercial AI services, a common constraint in university health research and in public agencies.","evidence":"University news article with faculty-reported early results; the 10x figure is a faculty statement without published method or baseline details.","architectureImplications":"On-premises DGX B200 systems with storage and networking let regulated data stay inside university-controlled infrastructure.","governanceImplications":"Local training on patient data still requires IRB, HIPAA and data-use agreement controls; owning hardware does not replace them.","securityPrivacyImplications":"Keeping protected health information off external AI services is the stated motivation; the article does not describe access controls or audit.","caveats":"Promotional university article; early anecdotal results; prior Cal Poly AI Factory coverage is already archived, and this item is filed only for the new early-adopter details.","streamIds":["nvidia"],"roles":{"sales":"Cal Poly's early users point to the strongest on-premises argument in higher education: protected data that cannot go to outside AI services. With research vice presidents and CIOs, ask which projects are stalled for that reason. The 10x eelgrass result is a faculty anecdote, useful for interest but not as a promise. A bounded engagement could inventory regulated-data AI demand and test whether shared on-premises GPU capacity would unblock it.","engineering":"Engineering should focus on the regulated-data workflow: how protected health data moves into the AI Factory, who can access it and how activity is logged. Four DGX B200 systems require scheduling, storage and network design that separates projects. Verify the 10x speed claim against a documented baseline before reusing it. Align controls with HIPAA, IRB and data-use agreements, since local hardware alone does not satisfy them.","delivery":"Delivery should set up project onboarding that checks data classification, IRB approval and access rules before regulated work starts. Define allocation policies for teaching versus research. Record early adopter outcomes with methods and baselines so later reports are credible. Assign an operational owner for the systems and plan support staffing, since faculty demand is likely to grow once early results circulate across campus."},"roleDetails":{"sales":{"customerProblem":"Research involving protected data stalls because it cannot use outside AI services.","stakeholders":["Vice president for research","CIO","IRB or research compliance office","Faculty PIs"],"discoveryQuestions":["Which projects are blocked by data restrictions?","What controls exist for regulated data on research compute?"],"valueHypothesis":"Shared on-premises GPU capacity could unblock regulated-data AI research under existing compliance controls.","potentialEngagement":"Inventory of regulated-data AI demand with a controls-readiness review.","unsupportedClaims":["The 10x speed-up as a general result","Compliance through hardware ownership alone"]},"engineering":{"fit":"Universities with health or other regulated research data.","architectureIntegration":"On-premises DGX B200 systems with segregated storage and project scheduling.","prerequisites":["Data classification and access controls","Project-level isolation","Audit logging"],"constraints":["Anecdotal results only","Capital and operations funding required"],"security":"Segregate projects, log access and align with HIPAA, IRB and data-use agreements.","proofOfValue":["Documented baseline for speed claims","Access-control audit for a regulated project"]},"delivery":{"work":"Onboard regulated-data projects through compliance checks and measure outcomes.","dependencies":["IRB and compliance office","Systems operations staff"],"ownership":"Research computing with research compliance oversight.","skillsAdoption":"Train faculty on secure data handling on shared systems.","governanceCheckpoints":["Data classification check at onboarding","IRB approval verified"],"acceptanceCriteria":["Proposed: regulated projects onboarded with documented controls","Proposed: outcomes reported with methods and baselines"],"risks":["Unauthorised access to protected data","Overstated early results","Capacity contention"]}},"retrievedAt":"2026-10-29T02:59:09Z","enrichedAt":"2026-10-29T03:06:28Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Gives students direct experience training models, framed around evaluating AI output and its limits.","procurementImplications":"Funded by the Noyce School's gift-backed budget; replication depends on capital funding and operations support.","operatingModelImplications":"An on-campus AI factory needs operations, allocation and support models across faculty users.","sourceVerification":{"openedUrl":"https://ceng.calpoly.edu/news/cal-poly-unveils-nvidia-powered-ai-factory-expand-ai-education-and-research","referenceExcerpt":"Because protected patient information cannot be sent to outside AI services, the AI Factory allows researchers to train models locally.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}