{"resourceId":"get-gyeongnam-ai-wildfire-cctv-79pct-false-detections-20260927","versions":[{"version":"external-be9b1ed0b9854871590aa53d73a1602f2530f42ff0c43e4c06885ccc7edea6d7","resource":{"id":"get-gyeongnam-ai-wildfire-cctv-79pct-false-detections-20260927","title":"Korea Forest Service data: Gyeongnam AI wildfire CCTV platform logged 79% error detections, 70% nationally","organization":"Korea Forest Service / Gyeongnam Province","sector":"National and provincial forest-fire surveillance","geography":"South Korea","publishedAt":"2026-09-27","publicationDate":"2026-09-27","eventDate":null,"sourceName":"Global Economic Times (Ko Yong-chul)","sourceLabel":"Global Economic Times report on Korea Forest Service ICT wildfire platform error-detection data","sourceUrl":"https://globaleconomictimes.kr/article/1065573633407777","evidenceClass":"independent-reporting","outcomeClass":"cautionary","topics":["infrastructure","governance-procurement","operating-model"],"finding":"Citing Korea Forest Service data compiled on the 23rd, the Sept. 27, 2026 report says Gyeongnam's AI CCTV wildfire platform made 106,061 detections last year; 22,140 were real smoke or flames and 83,921 were errors, a 79% error rate against a 70% national average. Glare, lens debris, clouds, fog and swaying trees caused misreads. Officials said accuracy falls short and upgrades continue.","sledRelevance":"Gives rare government-sourced, large-sample false-detection numbers for AI camera wildfire detection, a counterweight to U.S. and Canadian early-detection anecdotes.","evidence":"Agency statistics relayed by a newspaper; the underlying dataset was not opened. How errors are counted and whether humans filtered them before dispatch is unclear.","architectureImplications":"CCTV feeds analysed by AI in real time with human confirmation; error volume drives operator workload.","governanceImplications":"Shows why procurement should require reported false-positive rates and local tuning commitments.","securityPrivacyImplications":"Not discussed.","caveats":"Korean terrain, camera models and counting rules may differ; error share is per detection, not per dispatch. Published three weeks ago; newly relevant as Whistler and Los Altos Hills tout detection this week.","streamIds":["emergency-services"],"roles":{"sales":"This is the evidence to keep wildfire-detection pitches honest: in Gyeongnam, 79% of AI camera detections were errors. Use it to steer buyers toward contracts that measure false positives and fund human verification. It supports a discovery question, not a claim that U.S. systems perform the same.","engineering":"The platform runs real-time AI on CCTV with glare, debris, fog and moving trees causing misreads. Engineering should ask how any proposed system filters these locally, how errors are counted, and how many reach dispatch. Proof of value should include a per-camera false-positive log over a full season.","delivery":"Delivery must budget for human screeners and tuning, and set a target error rate with reporting. Ownership sits with the forest or fire agency running the platform. Acceptance criteria below are proposals."},"roleDetails":{"sales":{"customerProblem":"Buyers of AI wildfire cameras lack realistic false-alarm expectations.","stakeholders":["State forestry agency","Fire chiefs","Procurement office","Budget office"],"discoveryQuestions":["What false-positive rate would your screeners tolerate?","Who verifies alerts today?","Will the contract require error reporting?"],"valueHypothesis":"Contracts with error-rate reporting and tuning may avoid paying for detections that mostly create screening work.","potentialEngagement":"Procurement-requirements review for detection systems, including error metrics.","unsupportedClaims":["U.S. systems have 79% error rates","AI detection is useless","Specific vendor comparison"]},"engineering":{"fit":"Large-scale AI CCTV wildfire detection with human confirmation.","architectureIntegration":"Existing surveillance CCTV feeds an AI analysis layer that raises candidate detections for human review.","prerequisites":["Camera maintenance (lens cleaning)","Local training data","Error logging"],"constraints":["Korean context","Counting method unclear","Dataset not opened"],"security":"Not addressed.","proofOfValue":["Per-camera false-positive rates by cause","Share of detections reaching dispatch","Before/after rates after model updates"]},"delivery":{"work":"Track errors by cause, fund screening staff, and require vendor tuning against targets.","dependencies":["Camera vendor","AI vendor","Forestry operations centre"],"ownership":"Forest fire agency operations centre.","skillsAdoption":"Screeners learn common false-alarm patterns.","governanceCheckpoints":["Annual error-rate report","Contract renewal tied to accuracy","Budget review"],"acceptanceCriteria":["Error rate published annually","Target reduction set per year","Maintenance schedule for lens cleaning"],"risks":["Alert fatigue","Sunk-cost spending","Missed fires amid noise"]}},"retrievedAt":"2026-10-09T22:02:00-05:00","enrichedAt":"2026-10-09T22:02:00-05:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"High error volume shifts work to human screeners rather than removing it.","procurementImplications":"Report cites 20.25 billion won spent so far; spend without accuracy targets is the procurement lesson.","operatingModelImplications":"Human verification layer is essential when most AI detections are wrong.","sourceVerification":{"responseId":"curl-get-2026-10-09","openedUrl":"https://globaleconomictimes.kr/article/1065573633407777","referenceExcerpt":"The error detection rate was recorded at 79%, which is 9 percentage points higher than the national average (70%) last year.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}