{"work":{"id":"485486b1-a1a2-4cde-bdda-768930c403e6","openalex_id":null,"doi":null,"arxiv_id":"2502.18864","raw_key":null,"title":"Towards an AI co-scientist","authors":null,"authors_text":"Juraj Gottweis, Wei-Hung Weng, Alexander Daryin, Tao Tu, Anil Palepu, Petar Sirkovic","year":2025,"venue":"cs.AI","abstract":"Scientific discovery relies on scientists generating novel hypotheses that undergo rigorous experimental validation. To augment this process, we introduce an AI co-scientist, a multi-agent system built on Gemini 2.0. The AI co-scientist is intended to help uncover new, original knowledge and to formulate demonstrably novel research hypotheses and proposals, building upon prior evidence and aligned to scientist-provided research objectives and guidance. The system's design incorporates a generate, debate, and evolve approach to hypothesis generation, inspired by the scientific method and accelerated by scaling test-time compute. Key contributions include: (1) a multi-agent architecture with an asynchronous task execution framework for flexible compute scaling; (2) a tournament evolution process for self-improving hypotheses generation. Automated evaluations show continued benefits of test-time compute, improving hypothesis quality. While general purpose, we focus development and validation in three biomedical areas: drug repurposing, novel target discovery, and explaining mechanisms of bacterial evolution and anti-microbial resistance. For drug repurposing, the system proposes candidates with promising validation findings, including candidates for acute myeloid leukemia that show tumor inhibition in vitro at clinically applicable concentrations. For novel target discovery, the AI co-scientist proposed new epigenetic targets for liver fibrosis, validated by anti-fibrotic activity and liver cell regeneration in human hepatic organoids. Finally, the AI co-scientist recapitulated unpublished experimental results via a parallel in silico discovery of a novel gene transfer mechanism in bacterial evolution. These results, detailed in separate, co-timed reports, demonstrate the potential to augment biomedical and scientific discovery and usher an era of AI empowered scientists.","external_url":"https://arxiv.org/abs/2502.18864","cited_by_count":null,"metadata_source":"pith","metadata_fetched_at":"2026-05-14T23:08:14.671879+00:00","pith_arxiv_id":"2502.18864","created_at":"2026-05-08T21:14:12.379699+00:00","updated_at":"2026-05-14T23:08:14.671879+00:00","title_quality_ok":false,"display_title":"Towards an AI co-scientist","render_title":"Towards an AI co-scientist"},"hub":{"state":{"work_id":"485486b1-a1a2-4cde-bdda-768930c403e6","tier":"hub","tier_reason":"10+ Pith inbound or 1,000+ external citations","pith_inbound_count":42,"external_cited_by_count":null,"distinct_field_count":12,"first_pith_cited_at":"2025-03-12T17:35:03+00:00","last_pith_cited_at":"2026-05-12T00:24:30+00:00","author_build_status":"not_needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"not_needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-05-15T01:46:18.846452+00:00","tier_text":"hub"},"tier":"hub","role_counts":[{"context_role":"background","n":5},{"context_role":"other","n":1}],"polarity_counts":[{"context_polarity":"background","n":5},{"context_polarity":"unclear","n":1}],"runs":{"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-14T17:48:45.513160+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery","work_id":"56b6b58d-e73a-4317-896e-36ac5f84e957","shared_citers":23},{"title":"The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search","work_id":"fa04f346-ee20-4e9d-bf04-3ad3569a8ed1","shared_citers":11},{"title":"AlphaEvolve: A coding agent for scientific and algorithmic discovery","work_id":"76a0f850-d490-4e4f-ab98-8d25df82cd23","shared_citers":7},{"title":"Ai-researcher: Autonomous scientific innovation","work_id":"3845f0f0-08d4-4650-b390-6bfdd269f79a","shared_citers":6},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":6},{"title":"Deepscientist: Advancing frontier-pushing scientific findings progressively","work_id":"d68e01e1-6d49-4438-9cb6-114f102063a8","shared_citers":5},{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":5},{"title":"Landsness, Daniel L","work_id":"f21b1a4c-dda4-447b-a222-692f1ecf62dd","shared_citers":5},{"title":"DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines","work_id":"d490f594-f5fc-47b0-ae6a-6550e50fe095","shared_citers":4},{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":4},{"title":"Humanity's Last Exam","work_id":"59ea00d4-16a8-45e1-aafc-290a6f91d9f4","shared_citers":4},{"title":"Mle-bench: Evaluating machine learning agents on machine learning engineering","work_id":"a671e43f-ceab-49e7-adc3-473d802a97ca","shared_citers":4},{"title":"OpenHands: An Open Platform for AI Software Developers as Generalist Agents","work_id":"f1762ea0-e382-4f38-a28c-adc643789859","shared_citers":4},{"title":"2310.03302 , archivePrefix =","work_id":"5655b20a-fbb7-4a39-8605-d6e1d689895a","shared_citers":3},{"title":"Agent laboratory: Using llm agents as research assistants","work_id":"cb6e0bd7-2f2d-4052-875a-985ba5600340","shared_citers":3},{"title":"arXiv preprint arXiv:2509.23426 (2025)","work_id":"e515adbd-38fb-496f-ba70-997fe71e0366","shared_citers":3},{"title":"AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation","work_id":"92b7eb9c-c3d8-4518-a376-06fa15dd895b","shared_citers":3},{"title":"BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents","work_id":"25adb508-d97c-49d6-ae43-7a70c2478a34","shared_citers":3},{"title":"Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers","work_id":"95c4070f-48fa-44f6-bed8-a4b874c54eac","shared_citers":3},{"title":"Constitutional AI: Harmlessness from AI Feedback","work_id":"faaaa4e0-2676-4fac-a0b4-99aef10d2095","shared_citers":3},{"title":"Cycleresearcher: Improving automated research via automated review","work_id":"529c57f4-8402-4221-93f2-032f08d64085","shared_citers":3},{"title":"did-the-right-thing-happen","work_id":"c949e94f-fa16-46a7-b6f1-96613d49b7d0","shared_citers":3},{"title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","work_id":"008df105-2fdd-45d8-857a-8e35868aecb6","shared_citers":3},{"title":"Georgiev, J","work_id":"b3882729-74ba-4d80-9240-bfe03c45c33b","shared_citers":3}],"time_series":[{"n":3,"year":2025},{"n":37,"year":2026}],"dependency_candidates":[]},"error":null,"updated_at":"2026-05-14T17:49:05.034971+00:00"},"identity_refresh":{"job_type":"identity_refresh","status":"succeeded","result":{"items":[{"title":"Qwen3 Technical Report","outcome":"unchanged","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","resolver":"local_arxiv","confidence":0.98,"old_work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e"}],"counts":{"fixed":0,"merged":0,"unchanged":1,"quarantined":0,"needs_external_resolution":0},"errors":[],"attempted":1},"error":null,"updated_at":"2026-05-14T17:49:27.074341+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"Towards an AI co-scientist","claims":[{"claim_text":"Scientific discovery relies on scientists generating novel hypotheses that undergo rigorous experimental validation. To augment this process, we introduce an AI co-scientist, a multi-agent system built on Gemini 2.0. The AI co-scientist is intended to help uncover new, original knowledge and to formulate demonstrably novel research hypotheses and proposals, building upon prior evidence and aligned to scientist-provided research objectives and guidance. The system's design incorporates a generate, debate, and evolve approach to hypothesis generation, inspired by the scientific method and accele","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Towards an AI co-scientist because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-14T17:49:07.855109+00:00"}},"summary":{"title":"Towards an AI co-scientist","claims":[{"claim_text":"Scientific discovery relies on scientists generating novel hypotheses that undergo rigorous experimental validation. To augment this process, we introduce an AI co-scientist, a multi-agent system built on Gemini 2.0. The AI co-scientist is intended to help uncover new, original knowledge and to formulate demonstrably novel research hypotheses and proposals, building upon prior evidence and aligned to scientist-provided research objectives and guidance. The system's design incorporates a generate, debate, and evolve approach to hypothesis generation, inspired by the scientific method and accele","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Towards an AI co-scientist because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery","work_id":"56b6b58d-e73a-4317-896e-36ac5f84e957","shared_citers":23},{"title":"The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search","work_id":"fa04f346-ee20-4e9d-bf04-3ad3569a8ed1","shared_citers":11},{"title":"AlphaEvolve: A coding agent for scientific and algorithmic discovery","work_id":"76a0f850-d490-4e4f-ab98-8d25df82cd23","shared_citers":7},{"title":"Ai-researcher: Autonomous scientific innovation","work_id":"3845f0f0-08d4-4650-b390-6bfdd269f79a","shared_citers":6},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":6},{"title":"Deepscientist: Advancing frontier-pushing scientific findings progressively","work_id":"d68e01e1-6d49-4438-9cb6-114f102063a8","shared_citers":5},{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":5},{"title":"Landsness, Daniel L","work_id":"f21b1a4c-dda4-447b-a222-692f1ecf62dd","shared_citers":5},{"title":"DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines","work_id":"d490f594-f5fc-47b0-ae6a-6550e50fe095","shared_citers":4},{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":4},{"title":"Humanity's Last Exam","work_id":"59ea00d4-16a8-45e1-aafc-290a6f91d9f4","shared_citers":4},{"title":"Mle-bench: Evaluating machine learning agents on machine learning engineering","work_id":"a671e43f-ceab-49e7-adc3-473d802a97ca","shared_citers":4},{"title":"OpenHands: An Open Platform for AI Software Developers as Generalist Agents","work_id":"f1762ea0-e382-4f38-a28c-adc643789859","shared_citers":4},{"title":"2310.03302 , archivePrefix =","work_id":"5655b20a-fbb7-4a39-8605-d6e1d689895a","shared_citers":3},{"title":"Agent laboratory: Using llm agents as research assistants","work_id":"cb6e0bd7-2f2d-4052-875a-985ba5600340","shared_citers":3},{"title":"arXiv preprint arXiv:2509.23426 (2025)","work_id":"e515adbd-38fb-496f-ba70-997fe71e0366","shared_citers":3},{"title":"AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation","work_id":"92b7eb9c-c3d8-4518-a376-06fa15dd895b","shared_citers":3},{"title":"BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents","work_id":"25adb508-d97c-49d6-ae43-7a70c2478a34","shared_citers":3},{"title":"Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers","work_id":"95c4070f-48fa-44f6-bed8-a4b874c54eac","shared_citers":3},{"title":"Constitutional AI: Harmlessness from AI Feedback","work_id":"faaaa4e0-2676-4fac-a0b4-99aef10d2095","shared_citers":3},{"title":"Cycleresearcher: Improving automated research via automated review","work_id":"529c57f4-8402-4221-93f2-032f08d64085","shared_citers":3},{"title":"did-the-right-thing-happen","work_id":"c949e94f-fa16-46a7-b6f1-96613d49b7d0","shared_citers":3},{"title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","work_id":"008df105-2fdd-45d8-857a-8e35868aecb6","shared_citers":3},{"title":"Georgiev, J","work_id":"b3882729-74ba-4d80-9240-bfe03c45c33b","shared_citers":3}],"time_series":[{"n":3,"year":2025},{"n":37,"year":2026}],"dependency_candidates":[]},"authors":[]}}