The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery
By Chris Lu, Cong Lu, R. Lange, Jakob Foerster, Jeff Clune, David Ha. arXiv.org, 2024
957 citations recorded by the source corpus.
One of the grand challenges of artificial general intelligence is developing agents capable of conducting scientific research and discovering new knowledge. While frontier models have already been used as aides to human scientists, e.g. for brainstorming ideas, writing code, or prediction tasks, they still conduct only a small part of the scientific process. This paper presents the first comprehensive framework for fully automatic scientific discovery, enabling frontier large language models to perform research independently and communicate their findings. We introduce The AI Scientist, which generates novel research ideas, writes code, executes experiments, visualizes results, describes its findings by writing a full scientific paper, and then runs a simulated review process for evaluation. In principle, this process can be repeated to iteratively develop ideas in an open-ended fashion, acting like the human scientific community. We demonstrate its versatility by applying it to three distinct subfields of machine learning: diffusion modeling, transformer-based language modeling, and learning dynamics. Each idea is implemented and developed into a full paper at a cost of less than $15 per paper. To evaluate the generated papers, we design and validate an automated reviewer, which we show achieves near-human performance in evaluating paper scores. The AI Scientist can produce papers that exceed the acceptance threshold at a top machine learning conference as judged by our automated reviewer. This approach signifies the beginning of a new era in scientific discovery in machine learning: bringing the transformative benefits of AI agents to the entire research process of AI itself, and taking us closer to a world where endless affordable creativity and innovation can be unleashed on the world's most challenging problems. Our code is open-sourced at https://github.com/SakanaAI/AI-Scientist
Subfields: Information Systems and Management
View Paper: https://www.semanticscholar.org/paper/33161a5a9b5dcb635b5a97475e6a6209a69ada7d
Citations in the corpus
- Large Language Models for Information Retrieval: A Survey (2023, neutral): “Although some existing studies have used reinforcement learning to adjust the query rewriting process according to generation results [100], a substantial realm of research remains unexplored concerning the integration of ranking results.”
- Characterizing LLM scientific concept generation: A multi-dimensional measurement study (2026, unknown): “Several studies have explored LLM capabilities in scientific ideation and concept generation [1,25,26].”
- ResearchGPT: Benchmarking and Training LLMs for End-to-End Computer Science Research Workflows (2025, unknown): “Recent initiatives like ResearchBench (Liu et al., 2025) and MOOSE-Chem2 (Yang et al., 2025) examine hypothesis generation and fine-grained scientific reasoning, while systems such as Agent Laboratory (Schmidgall et al., 2025) and the AI Scientist (Lu et al., 2024) push toward autonomous end-to-end ”
- Bloom-Eval: A Hierarchical Evaluation Benchmark for Automatic Survey Generation Based on Bloom's Taxonomy (2026, unknown): “A primary objective of artificial intelligence (AI) for science is to evolve AI from a mere information processor into a genuine knowledge generator (Lu et al., 2024; Li et al., 2025; Yuan et al., 2025).”
- Fusing Multi‐ and Hyperspectral Satellite Data for Harmful Algal Bloom Monitoring With Self‐Supervised and Hierarchical Deep Learning (2025, unknown): “To accelerate exploration, models can be deployed for collaborative, open-ended discovery, allowing us to rapidly sift through extensive data volumes and surface new, atypical, or noteworthy observations for further investigation (Zhang et al., 2023; Lu et al., 2024).”