{"resourceId":"uchicago-living-science-ai-agents-reproduce-update-studies-20261102","versions":[{"version":"external-206f8af3f42c0b62170fd161f8b94df301d295c4515b9926f670b08611a15046","resource":{"id":"uchicago-living-science-ai-agents-reproduce-update-studies-20261102","title":"UChicago Living Science uses AI agents to reproduce influential papers and rerun them on newer data, starting with a minimum-wage study","organization":"University of Chicago Department of Computer Science (Chenhao Tan, Evan Rose); SAI Labs","sector":"University research (AI-assisted reproduction and updating of published studies)","geography":"United States","publishedAt":"2 November 2026 (UChicago CS News)","publicationDate":"2026-11-02","eventDate":null,"sourceName":"UChicago Department of Computer Science","sourceLabel":"UChicago CS News, \"Keeping Science Alive: A University of Chicago Project is Using AI to Update Old Science,\" 2 Nov 2026","sourceUrl":"https://computerscience.uchicago.edu/news/keeping-science-alive-a-university-of-chicago-project-is-using-ai-to-update-old-science/","evidenceClass":"academic-research","outcomeClass":"emerging","topics":["developers-agents","knowledge-work","operating-model"],"finding":"Living Science, led by associate professor Chenhao Tan, uses AI agents to reproduce an influential paper from its original materials and rerun the analysis on newer data, publishing updated findings with code and data. First topics are in economics, chosen by collaborator Evan Rose; five AI safety papers are in progress. One early report revisited a 2019 minimum-wage study: the team first matched the published results, then expanded the sample from about 138 to 236 state minimum-wage increases through 2024. The original conclusion held (affected workers earned more, by less than first found, with no clear change in employment), and the report notes parts the team could not fully reproduce. Tan framed the project as constructive rather than a weapon to criticize existing work and invites feedback from original authors.","sledRelevance":"Offers research offices and libraries a concrete model for using agents on reproducibility and evidence maintenance, an integrity function rather than new discovery. It also raises questions about credit, author consultation and how institutions present revised findings to policymakers.","evidence":"Read the full university news story. It describes one completed report; agent accuracy, human review effort, cost and failure rates are not reported, and the reports themselves were not opened. Departmental news, not an independent evaluation.","architectureImplications":"Agents operate on original replication materials plus newer public data; outputs are published with code and data for external checking.","governanceImplications":"Raises norms for re-analysing others' work: author feedback, transparency and avoiding adversarial framing.","securityPrivacyImplications":"Not addressed; first topics use public economic data.","caveats":"University self-report of an early-stage project with one described result; how much the agents did versus humans is unclear. Economics with public data may not transfer to fields with restricted data or wet-lab work.","streamIds":["research"],"roles":{"sales":"Lead with reproducibility as a research-integrity service. UChicago said on 2 November that its Living Science project uses AI agents to reproduce influential papers and rerun them on newer data; a minimum-wage study expanded from about 138 to 236 state increases and its conclusion held. Research integrity officers, libraries and research computing leaders are the audience. Offer a bounded pilot design for agent-assisted reproduction of a few institutional papers. Do not claim agents reproduce studies reliably; no accuracy data was published.","engineering":"Frame fit around reproducible agent workflows on public replication packages. Living Science has agents reproduce results from original materials, then rerun analyses with newer data, publishing code and data for checking. Prerequisites are accessible replication packages, versioned data pipelines and a sandboxed compute environment. Validate by first matching published results exactly, logging which steps could not be reproduced, and having a domain expert review every updated analysis before release, as the project reports partial reproduction gaps.","delivery":"Treat delivery as a small reproduction service. Work covers selecting papers, securing replication materials, running agents, expert review, author consultation and publishing results. Ownership sits with a research integrity office or library data services, with faculty domain reviewers. Governance checkpoints cover author notification and how revised findings are communicated. Acceptance criteria are proposals, such as exact reproduction of published estimates before any update, not results UChicago reported."},"roleDetails":{"sales":{"customerProblem":"Influential findings are rarely updated, and institutions lack capacity to check whether cited results still hold.","stakeholders":["Research integrity officer","University librarian or data services lead","Vice president for research","Faculty authors of high-impact papers"],"discoveryQuestions":["Which institutional papers drive policy citations?","Are replication packages available for them?","Who would review agent-updated analyses?"],"valueHypothesis":"Agents may cut the labour of reproducing and updating public-data studies, if expert review stays in the loop.","potentialEngagement":"A pilot design to reproduce two or three public-data papers with agents and expert review.","unsupportedClaims":["Agents reproduce studies reliably","Findings can be updated without expert review","Results apply to restricted-data fields"]},"engineering":{"fit":"Fits computational studies with public data and replication packages, starting in economics.","architectureIntegration":"Agent runs in a sandbox against original code and data, then newer data; outputs versioned and published with code.","prerequisites":["Replication packages","Versioned data pipelines","Sandboxed compute"],"constraints":["Restricted or human-subjects data","Incomplete original materials","No published agent accuracy data"],"security":"Keep agents in isolated environments without credentials to restricted datasets.","proofOfValue":["Exact match to published estimates","Log of unreproduced steps","Expert sign-off on updated analysis"]},"delivery":{"work":"Select papers, gather materials, run agents, review results, consult authors and publish.","dependencies":["Replication materials","Domain reviewers","Compute environment"],"ownership":"Research integrity office or library data services with faculty reviewers.","skillsAdoption":"Staff learn agent supervision and reproduction review.","governanceCheckpoints":["Author notification before publication","Expert review of every update","Communication plan for changed findings"],"acceptanceCriteria":["Published estimates reproduced before update (proposal)","All deviations documented","Author feedback recorded"],"risks":["Silent agent analysis errors","Reputational conflict with original authors","Over-reading revised findings"]}},"retrievedAt":"2026-11-02T22:07:30-06:00","enrichedAt":"2026-11-02T22:07:32-06:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Treats checking old results as research work that careers rarely reward; implies new staff or student roles in maintaining findings.","procurementImplications":"Not addressed; tools are being built by the lab and SAI Labs.","operatingModelImplications":"Suggests institutions could run continuous reproduction and updating of high-impact papers as shared infrastructure.","peerReview":"unknown","sourceVerification":{"openedUrl":"https://computerscience.uchicago.edu/news/keeping-science-alive-a-university-of-chicago-project-is-using-ai-to-update-old-science/","referenceExcerpt":"bringing the number of state minimum wage increases studied to 236, up from about 138, and carried the analysis through 2024.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}