{"resourceId":"guardian-uk-ai-clog-public-bodies-complaints-foi-surge-20261108","versions":[{"version":"external-7b8d835df5d9f3e935f1bfd6b6a3fa2d2bfc9446075c0076c9e1a3f635809057","resource":{"id":"guardian-uk-ai-clog-public-bodies-complaints-foi-surge-20261108","title":"'AI clog': UK councils, MPs' offices and other public bodies report surging, far longer AI-generated complaints and FOI requests; local-authority complaints now often 10 times longer","organization":"Multiple UK public bodies (local authorities, Lawyers in Local Government, ICO, TfL, MPs' offices)","sector":"Complaints handling and public records (FOI) requests","geography":"United Kingdom","publishedAt":"2026-11-08 (last modified 2026-11-08)","publicationDate":"2026-11-08","eventDate":null,"sourceName":"The Guardian","sourceLabel":"'AI clog': UK public bodies battle onslaught of chatbot-generated complaints","sourceUrl":"https://www.theguardian.com/technology/2026/nov/08/ai-clog-public-bodies-local-authorities-onslaught-chatbot-generated-correspondence","evidenceClass":"independent-reporting","outcomeClass":"cautionary","topics":["operating-model","governance-procurement","knowledge-work"],"finding":"The Guardian (Nov. 8, 2026) reports UK public bodies swamped by AI-generated complaints, letters and requests. Lawyers in Local Government's chief executive says local-authority complaints, once about two pages, are now mostly 10 times longer, with one about 200 pages, and AI drafts cite case law from other jurisdictions and occasionally hallucinate. Enfield's council leader reports a significant uptick in AI-assisted parking appeals. The ICO found many authorities received more requests in 2025 and says AI-generated requests can be long, repetitive and harder to interpret, yet still need human answers. TfL received a record 4,769 FOI requests in 2025-26 and notes more AI-like appeals. Responses discussed include ombudsman word limits and templates, Wiltshire exploring AI summaries and draft replies, and researchers' 'agentic flooding' framing.","sledRelevance":"Public records and complaint intake are CE functions; this is the broadest current account of AI-generated demand on them, extending archived single-jurisdiction items (Edinburgh, Converse County, St. Paul, PA OOR) with cross-sector evidence and emerging controls. Published Nov. 8, after the last completed CE run.","evidence":"National newspaper reporting with named practitioners and cited ICO and TfL figures; mostly anecdotal and practitioner estimates rather than systematic measurement.","architectureImplications":"Pressure toward structured intake (templates, word limits) and AI-assisted triage or summarisation of inbound correspondence, as Wiltshire is exploring.","governanceImplications":"Statutory response deadlines still apply to AI-generated requests; ICO says they must be answered by humans; limits on submissions raise access-to-justice and fairness questions.","securityPrivacyImplications":"Using AI to summarise complainant correspondence raises personal-data handling questions; UK civil-service guidance cited advises not inputting personal data and keeping human final decisions.","caveats":"UK-wide and cross-sector (includes employment tribunals and MPs' offices beyond SLED analogues); '10 times longer' is a practitioner estimate; detection of AI authorship is inferred from length and style; no U.S. data.","streamIds":["citizen-engagement"],"roles":{"sales":"Sales framing: AI-generated requests are a demand-side problem for records and complaint teams, not an AI adoption story. The Guardian reports local-authority complaints now often ten times longer and a record 4,769 FOI requests at TfL. Discovery should ask how request length, volume and on-time rates have moved since 2024 and which intake controls are lawful. A bounded engagement is an intake and triage assessment. Do not claim AI triage will restore deadlines; no evidence is offered.","engineering":"Engineering framing: the controls discussed are structural intake changes, such as templates and word limits, and AI summarisation of inbound correspondence, which Wiltshire is exploring. Fit work is classifying requests, extracting the actual ask and flagging duplicates while keeping humans on responses, as the ICO requires. Prerequisites are case-management integration and personal-data rules. Proof of value should compare handling time and accuracy of extracted requests on a sample of real AI-length submissions.","delivery":"Delivery framing: records and complaints teams need measurement before tools. Councils' complaint and FOI officers own the response, under statutory deadlines that do not change for AI-written requests. Acceptance criteria below are proposals: tracked request length and on-time rates, a human sign-off on every response, and an equity review before any submission limits. Pilot summarisation on a narrow request type, and keep assisted-drafting residents in mind when tightening intake."},"roleDetails":{"sales":{"customerProblem":"Complaint and FOI teams overwhelmed by long, AI-generated submissions under fixed statutory deadlines and budgets.","stakeholders":["FOI/records officers","Complaints teams","Monitoring officers and legal","Elected members"],"discoveryQuestions":["How have request length and on-time rates changed since 2024?","Which intake limits are lawful and fair?","Where could AI summarisation safely assist staff?"],"valueHypothesis":"Better intake structure and assistive triage may protect deadlines; impact unmeasured.","potentialEngagement":"An intake, triage and measurement assessment for records and complaints teams.","unsupportedClaims":["That AI triage restores on-time rates","Precise prevalence of AI-written requests","Transfer to U.S. law without review"]},"engineering":{"fit":"Fits agencies with statutory records or complaint deadlines and rising request complexity.","architectureIntegration":"Structured intake forms and optional AI summarisation feeding existing case management, with human-drafted responses.","prerequisites":["Case-management integration","Request classification scheme","Personal-data handling rules"],"constraints":["Statutory deadlines unchanged","Humans must answer","Restrictions may limit access"],"security":"Keep complainant personal data out of unapproved tools and log any AI summarisation.","proofOfValue":["Handling time on AI-length sample","Accuracy of extracted request","Duplicate detection rate"]},"delivery":{"work":"Measure the load, adjust intake and pilot assistive triage while preserving human responses.","dependencies":["Case-management system","Legal advice on limits","Staff capacity"],"ownership":"Information governance or complaints lead in each authority.","skillsAdoption":"Train officers to clarify scope with requesters and to use summarisation with verification.","governanceCheckpoints":["Legal review of intake limits","Equity review for assisted requesters","Human sign-off on every response"],"acceptanceCriteria":["Request length and on-time rate tracked monthly","Every response human-approved","Pilot handling-time change measured"],"risks":["Staff burnout","Restrictions seen as blocking access","Machine-to-machine escalation"]}},"retrievedAt":"2026-11-09T22:09:07-06:00","enrichedAt":"2026-11-09T22:15:04-06:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Caseworkers report exhaustion and unpaid hours; an MP's office notes AI also helps people who would struggle to draft correspondence, a key equity trade-off for any restriction.","procurementImplications":"Potential demand for triage and summarisation tools, with cost and 'human touch' concerns noted.","operatingModelImplications":"Complaint and FOI teams face longer, more complex inputs without extra funding; ombudsmen are reshaping intake channels to bound submissions.","sourceVerification":{"openedUrl":"https://www.theguardian.com/technology/2026/nov/08/ai-clog-public-bodies-local-authorities-onslaught-chatbot-generated-correspondence","referenceExcerpt":"The majority are now 10 times longer, says Evans.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}