{"resourceId":"openai-lifeval-everyday-tasks-government-services-eval-20261102","versions":[{"version":"external-630b23fd367c397f0c15d9d08000a290e0756ea04658d7e2d524064345b05442","resource":{"id":"openai-lifeval-everyday-tasks-government-services-eval-20261102","title":"OpenAI introduces LifeVal, a vendor evaluation of agentic everyday tasks including accessing government services","organization":"OpenAI","sector":"State and local government services; K-12 enrollment (citizen-facing)","geography":"United States and international","publishedAt":"November 2, 2026","publicationDate":"2026-11-02","eventDate":"2026-11-02","sourceName":"OpenAI","sourceLabel":"OpenAI","sourceUrl":"https://openai.com/index/lifeval","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["knowledge-work","developers-agents","accessibility-workforce"],"finding":"OpenAI describes LifeVal, an internal evaluation of everyday tasks built from the American Time Use Survey, the Multinational Time Use Study and a 3,995-respondent survey, spanning areas including accessing government services and education. Example task scores include 93.1% (kindergarten enrollment), 98.0% (flood volunteer coordination) and 100% (finding a home and adding a county deed to a records cart).","sledRelevance":"Indicates OpenAI is optimizing ChatGPT agents to act on resident-facing government and school workflows (enrollment, county records, emergency prep), which could change how residents interact with public portals.","evidence":"Vendor-built and vendor-scored benchmark; the page shows selected examples, not full methodology, grading rubric, task counts, environments, or independent replication.","architectureImplications":"Agentic browser/computer-use interactions with public portals imply bot traffic and automated form submissions on government sites.","governanceImplications":"Agencies may need policies on AI agents acting for residents (authorization, identity, liability).","securityPrivacyImplications":"Agents transacting on county records or enrollment systems raise identity-verification and data-handling questions.","caveats":"Vendor claim; scores are on curated tasks with unknown grading; real government portals, accessibility, and failure modes were not reported; no SLED agency involvement.","streamIds":["openai"],"roles":{"sales":"Interpretation: LifeVal signals that resident-facing agents are a design target, so digital services leaders should anticipate AI agents completing enrollment and records tasks for residents. Conversations should explore readiness of public portals for agent traffic, not cite the scores as performance proof, since they are vendor-reported on selected examples without methodology.","engineering":"Interpretation: engineering relevance is how government portals handle automated browser and computer-use agents: authentication, CAPTCHA, rate limits, structured APIs and audit trails. The source gives no task environment details, so any claim of reliability on real county or school systems must be validated independently in sandboxed tests.","delivery":"Interpretation: delivery work for agencies is policy and testing: define acceptable agent use on portals, monitor automated transactions, and test common resident journeys with agents. Owners are citizen digital services teams. Risks include erroneous agent-submitted transactions and unclear liability; acceptance criteria should be measured error rates in agency-run tests."},"roleDetails":{"sales":{"customerProblem":"Residents increasingly may use AI agents to interact with government and school portals not designed for them.","stakeholders":["Digital services directors","County records offices","School district enrollment offices"],"discoveryQuestions":["How do portals detect and handle AI agent sessions?","What transactions require human verification?"],"valueHypothesis":"Preparing portals for agent traffic may reduce errors and support residents.","potentialEngagement":"A bounded test of two resident journeys using AI agents in a sandbox.","unsupportedClaims":["Benchmark scores reflect real portal performance","Agents are safe for unsupervised transactions"]},"engineering":{"fit":"Relevant to citizen-facing services as an emerging pattern; vendor eval only.","architectureIntegration":"Agents use browser use, computer use and connectors per vendor.","prerequisites":["Sandbox portal environments","Agent usage policy"],"constraints":["No methodology published","Varied portal authentication"],"security":"Identity verification and authorization for agent-submitted transactions must be assessed.","proofOfValue":["Run agent on representative portal tasks and measure completion and error rates"]},"delivery":{"work":"Agent-readiness testing and policy for public portals.","dependencies":["Portal vendors","Legal guidance on delegated access"],"ownership":"Agency digital services team.","skillsAdoption":"Staff testing of agentic tools.","governanceCheckpoints":["Policy on delegated agent access","Privacy review"],"acceptanceCriteria":["Documented completion and error rates for tested journeys"],"risks":["Erroneous transactions","Unclear liability","Accessibility gaps"]}},"retrievedAt":"2026-11-03T04:07:46Z","enrichedAt":"2026-11-03T04:07:46Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Could aid residents navigating complex services, but accessibility outcomes were not measured.","procurementImplications":"No procurement content.","operatingModelImplications":"Citizen-engagement teams may see agent-mediated traffic.","sourceVerification":{"openedUrl":"https://openai.com/index/lifeval","referenceExcerpt":"we expanded on these existing surveys with a new demographically representative survey of 3,995 respondents.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}