{"resourceId":"slu-copilot-chat-default-staff-ai-procurement-rules-20261106","versions":[{"version":"external-ac9b4b98e138eec81a38607630cdda2ed707e37e4e4f3b5ceae1880147562aed","resource":{"id":"slu-copilot-chat-default-staff-ai-procurement-rules-20261106","title":"Swedish University of Agricultural Sciences makes licence-included Copilot Chat the default staff AI tool and puts AI purchases under IT procurement rules","organization":"Swedish University of Agricultural Sciences (SLU), university administration and IT Support","sector":"Higher education administration and IT","geography":"Sweden","publishedAt":"6 November 2026 (staff intranet news)","publicationDate":"2026-11-06","eventDate":null,"sourceName":"SLU Medarbetarwebben (staff intranet)","sourceLabel":"Institutional staff guidance","sourceUrl":"https://internt.slu.se/en/news-originals/2026/11/ai-at-slu/","evidenceClass":"standards-guidance","outcomeClass":"emerging","topics":["knowledge-work","governance-procurement","data-security"],"finding":"SLU's administration announced it is developing a generative AI strategy and, meanwhile, introducing Microsoft Copilot Chat, already included in its Microsoft licence, as the shared AI tool for general work tasks such as drafting, summarising, translation and brainstorming. Staff are told to use it instead of buying their own AI services, because many small purchases make it harder to protect information and track costs; AI purchases fall under IT procurement rules. A suitable/do-not-enter table excludes passwords, tenders in ongoing procurements, confidential research collaboration data and sensitive personal data. Staff must review outputs and not use AI for decisions such as grading exams or screening applications. Guidelines and training are promised by end of November; solutions for sensitive data later.","sledRelevance":"Shows a public university consolidating shadow AI spending onto a licence-included tool with explicit data-entry rules while a strategy is built; the same consolidation question faces U.S. campuses with departmental AI subscriptions.","evidence":"Primary institutional announcement; policy and intent only, no usage or outcome data.","architectureImplications":"Web-grounded Copilot Chat that only sees typed or uploaded content; no integration with mail or files; more integrated solutions deferred.","governanceImplications":"Interim guidance ahead of a full strategy; purchases routed through IT procurement; decision-making uses such as application screening prohibited.","securityPrivacyImplications":"Under SLU cloud rules, classified information and sensitive personal data generally may not be handled in cloud services, including AI; personal accounts discouraged because SLU lacks control.","caveats":"Announcement only; Swedish public-sector and GDPR context; no adoption, cost or compliance measures; model family and features may change.","streamIds":["campus-operations"],"roles":{"sales":"Sales framing: SLU is choosing to consolidate rather than buy, telling staff to stop purchasing their own AI services and use the Copilot Chat it already licenses. Interpretation: the near-term opportunity is helping institutions inventory departmental AI spend and define what moves to the default tool, not selling a new assistant. Evidence supports a governance and consolidation conversation; it says nothing about productivity or cost outcomes.","engineering":"Engineering framing: the tool is deliberately limited to typed or uploaded content with web search, and sensitive data is excluded under cloud rules. Pre-sales should map which data classes each approved tool may process and what a later 'sensitive-data' solution would need, such as tenancy, residency and logging. Copilot Chat's model choices change over time, so documented controls must track the vendor's configuration.","delivery":"Delivery framing: SLU has guidance now and training promised for end of November. Delivery work is turning the suitable/do-not-enter table into role-based training, setting up a request path for exceptions through IT procurement, and measuring whether departmental AI subscriptions actually decline. Keep the decision-use prohibition, such as application screening, visible in admissions and HR training."},"roleDetails":{"sales":{"customerProblem":"Departments buy many small AI services, fragmenting information protection and cost tracking.","stakeholders":["University director/administration","IT Support","Procurement","Data protection officer"],"discoveryQuestions":["How many departmental AI subscriptions exist today?","Which data classes may the default tool process?","What is the exception path for specialist AI needs?"],"valueHypothesis":"Consolidating on a licence-included tool may reduce untracked spend and data exposure; SLU reports no figures.","potentialEngagement":"An AI spend and data-class inventory with a consolidation and exception policy.","unsupportedClaims":["Measured cost savings","Productivity gains","Suitability for sensitive data"]},"engineering":{"fit":"Institutions already licensed for Microsoft 365 with no enterprise AI contract yet.","architectureIntegration":"Standalone web-grounded chat with organisational sign-in; no access to mail or files unless uploaded.","prerequisites":["Organisational identity sign-in","Data classification rules","Procurement policy covering AI"],"constraints":["Cloud rules bar sensitive personal data","GDPR and Swedish public-sector context","No integrated data access"],"security":"Enforce do-not-enter classes through training and, where possible, DLP; discourage personal accounts.","proofOfValue":["Count of departmental AI subscriptions before and after","Staff completion of AI guidance training","DLP or incident reports for prohibited data"]},"delivery":{"work":"Publish role-based guidance, run training, set up exception requests via IT procurement and track consolidation.","dependencies":["End-of-November guidelines","Procurement rule updates","Support desk capacity"],"ownership":"University administration with IT Support and procurement.","skillsAdoption":"Training on suitable uses, output review and prohibited decision uses.","governanceCheckpoints":["Strategy approval","Exception requests reviewed by IT procurement","Review before enabling sensitive-data solutions"],"acceptanceCriteria":["Proposed: inventory of AI subscriptions completed","Proposed: guidance training offered to all staff","Proposed: documented exception decisions"],"risks":["Staff keep personal accounts","Default tool insufficient for specialist needs","Guidance lags vendor model changes"]}},"retrievedAt":"2026-11-07T21:49:10-06:00","enrichedAt":"2026-11-07T22:00:00-06:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Training programmes are promised but not yet available; staff carry responsibility for reviewing outputs.","procurementImplications":"Licence-included default to curb fragmented small AI purchases; exceptions only for specific needs via IT procurement.","operatingModelImplications":"A single shared tool plus central support mailbox as the interim operating model.","sourceVerification":{"openedUrl":"https://internt.slu.se/en/news-originals/2026/11/ai-at-slu/","referenceExcerpt":"Making lots of small purchases of different services makes it harder to protect information and keep track of costs.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}