{"resourceId":"aarhus-university-arts-student-information-chatbot-pilot-20261026","versions":[{"version":"external-e08982634d40d97bbf7ffb8782a244eb578ec14f639ac56bba001ba20cdb1dc9","resource":{"id":"aarhus-university-arts-student-information-chatbot-pilot-20261026","title":"Aarhus University pilots a chatbot for Arts students that answers only from the faculty student-information site and refers unclear questions to staff","organization":"Aarhus University, Faculty of Arts","sector":"Higher education student services and academic advising information","geography":"Aarhus, Denmark","publishedAt":"Aarhus University staff news, 26 October 2026","publicationDate":"2026-10-26","eventDate":"2026-10-26","sourceName":"Aarhus University (staff news)","sourceLabel":"Institutional announcement","sourceUrl":"https://medarbejdere.au.dk/en/news-articles/news/artikel/ny-chatbot-kan-hjaelpe-dig-med-at-faa-svar-paa-dine-spoergsmaal-om-dit-studie","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["knowledge-work","governance-procurement"],"finding":"From 26 October 2026, Aarhus University Arts students can use a chatbot on the Arts student-information site for questions about teaching, exams and other study matters. It is configured to answer exclusively from information on that site; when no clear answer is found, it always directs students to staff such as student counsellors and the Arts Studies Administration. Students can rate answers. The university describes it as a pilot to investigate whether AI can help students navigate existing information on rules, regulations and deadlines.","sledRelevance":"A scoped student-information assistant with explicit human referral, launched as a pilot with feedback collection; a common entry point for advising automation.","evidence":"Institutional announcement only; no outcomes yet.","architectureImplications":"Retrieval restricted to one faculty website, which bounds answers but depends on that content being current and complete.","governanceImplications":"Pilot status and a referral rule to human staff are stated; evaluation criteria are not published.","securityPrivacyImplications":"Not discussed; feedback and question logs may contain student information.","caveats":"No accuracy, usage or satisfaction data; Danish university context; vendor and model not named.","streamIds":["student-success"],"roles":{"sales":"Sales framing: this is a scoped, low-risk starting point for student-services AI: answers only from an official faculty site, with automatic referral to counsellors when unsure. It suits conversations with registrars and advising leaders who want help with rules and deadlines questions without automating advice. A bounded engagement could scope a similar pilot with clear evaluation measures. Do not claim reduced counsellor workload or better student outcomes; Aarhus reports none yet.","engineering":"Engineering framing: the architecture is retrieval restricted to a single faculty web domain, with a fallback that routes to named staff roles. Assess content quality and freshness of the source site, how the bot detects that no clear answer exists, and how ratings are captured. Proof of value should test answer accuracy against a question bank of exam and deadline rules and measure how often referrals occur and are appropriate.","delivery":"Delivery framing: the operational owner is effectively whoever maintains the student-information site, because answer quality depends on it. Delivery would set up content review cycles, define referral routes to counsellors and administration, collect and triage student ratings, and report pilot results to faculty leadership. Agree in advance what accuracy and referral performance would justify continuing, since the announcement states no evaluation criteria."},"roleDetails":{"sales":{"customerProblem":"Students struggle to find rules, deadlines and exam information spread across faculty websites.","stakeholders":["Faculty dean or vice dean for education","Studies administration","Student counsellors","Web content team"],"discoveryQuestions":["Which student questions take most counsellor and administrator time?","Is your student-information content current enough to ground a chatbot?","What would make you continue or end a pilot?"],"valueHypothesis":"Interpretation: a site-grounded assistant with staff referral may help students find existing information faster, if accuracy holds.","potentialEngagement":"A scoped pilot with a question-bank accuracy test and referral tracking.","unsupportedClaims":["Reduced counsellor workload","Improved retention or satisfaction","Accuracy levels, which are unreported"]},"engineering":{"fit":"Information-seeking questions answerable from official published content.","architectureIntegration":"Chat widget on the faculty student site with retrieval limited to that site and referral to staff roles.","prerequisites":["Current, well-structured student-information content","Defined referral contacts","Feedback capture and logging"],"constraints":["Answers limited to one site","No personalised record access","Danish institutional context"],"security":"Review whether question logs and feedback store personal data and apply retention limits.","proofOfValue":["Accuracy test on exam and deadline questions","Referral rate and appropriateness","Student ratings over the pilot period"]},"delivery":{"work":"Run the pilot with content maintenance, referral routing and feedback triage, then report results.","dependencies":["Web content owners","Studies administration capacity","Vendor or internal technical support"],"ownership":"Faculty studies administration with the web content team.","skillsAdoption":"Content editors trained to write answer-ready pages; staff briefed on handling referrals.","governanceCheckpoints":["Pre-launch accuracy check","Monthly feedback review","End-of-pilot decision against agreed criteria"],"acceptanceCriteria":["Proposed: agreed accuracy threshold on a rules question bank","Proposed: unanswered questions routed to staff in all test cases","Proposed: content errors corrected within a set time"],"risks":["Outdated content producing wrong deadlines","Students treating answers as official rulings","Low use limiting evaluation"]}},"retrievedAt":"2026-10-27T03:21:12Z","enrichedAt":"2026-10-27T03:25:47Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Referral to counsellors and administration keeps staff in the loop; workload effects unknown.","procurementImplications":"Not disclosed whether built or bought.","operatingModelImplications":"Interpretation: content owners for the student site become de facto owners of chatbot answer quality.","sourceVerification":{"openedUrl":"https://medarbejdere.au.dk/en/news-articles/news/artikel/ny-chatbot-kan-hjaelpe-dig-med-at-faa-svar-paa-dine-spoergsmaal-om-dit-studie","referenceExcerpt":"the chatbot will always direct you to the appropriate staff members, such as student counsellors, the Arts Studies Administration, and others who can help","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}