{"slug":"ai-for-science","topic":"ai-for-science","title":"AI for Science: State of the Field (October 2026)","summary":"This dossier maps AI for science as of October 2026, covering machine-learned interatomic potentials, computational materials discovery, autonomous laboratories and agentic research systems. The main takeaway is that computational candidate generation is now abundant, while experimental validation, benchmark design and reproducibility appear to be the main constraints in the sources reviewed. Protein structure prediction and weather forecasting are not covered in depth in this edition.","version":1,"published_at":"2026-10-30","updated_at":"2026-10-30","stats":{"words":3240,"citations":32,"claims":27,"timeline":19,"benchmarks":5,"open_problems":6,"key_papers":14,"glossary":15},"verification":{"status":"verified","method":"independent AI fact-check against cited sources (evidence pack + web checks); automated citation resolution","claims_checked":96,"issues_fixed":18,"verified_at":"2026-10-30"},"ai_assisted":true,"price":{"usd":"0.20","protocol":"x402","network":"eip155:84532","asset":"USDC"},"url":"https://space-kitty.com/dossiers/ai-for-science","preview_url":"https://space-kitty.com/api/v1/dossiers/ai-for-science/preview","purchase_url":"https://space-kitty.com/api/v1/dossiers/ai-for-science","toc":[{"level":2,"text":"Executive summary","anchor":"executive-summary"},{"level":2,"text":"Background and core concepts","anchor":"background-and-core-concepts"},{"level":2,"text":"Timeline of key milestones","anchor":"timeline-of-key-milestones"},{"level":2,"text":"State of the art","anchor":"state-of-the-art"},{"level":3,"text":"Learned interatomic potentials","anchor":"learned-interatomic-potentials"},{"level":3,"text":"Materials discovery and screening","anchor":"materials-discovery-and-screening"},{"level":3,"text":"Autonomous laboratories","anchor":"autonomous-laboratories"},{"level":3,"text":"Agentic and language-model systems","anchor":"agentic-and-language-model-systems"},{"level":3,"text":"Quantitative benchmarks","anchor":"quantitative-benchmarks"},{"level":2,"text":"Open problems and active debates","anchor":"open-problems-and-active-debates"},{"level":2,"text":"Major programs and players","anchor":"major-programs-and-players"},{"level":2,"text":"Outlook: what to watch (next 12-18 months)","anchor":"outlook-what-to-watch-next-12-18-months"},{"level":2,"text":"Key papers","anchor":"key-papers"},{"level":2,"text":"Glossary","anchor":"glossary"}],"sample_claims":["Universal machine-learned interatomic potentials covering broad element sets are now an established model class.","M3GNet combines a graph neural network with explicit three-body interactions and was designed to cover the periodic table."],"includes":["markdown","claims","timeline","benchmarks","open_problems","key_papers","glossary","bibliography"],"citation_count":32,"how_to_buy":"GET https://space-kitty.com/api/v1/dossiers/ai-for-science with an x402 client ($0.20, USDC on Base Sepolia (testnet))."}