How to build an AI Scientist: first peer-reviewed paper spills the secrets

2026-03-25 · Source: Machine learning : nature.com subject feeds · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation, AI for Scientific Research · Depth: Advanced, quick

Summary

AI Scientist, an autonomous research AI tool developed by Sakana AI, has undergone peer review, with a paper detailing its capabilities published in *Nature*. Initially unveiled in August 2024, the system aims to automate the entire scientific process, from idea generation and hypothesis testing to code execution and paper writing. The tool, built on existing large language models like GPT-4o or Claude Sonnet 4, generated three research papers, one of which was accepted by peer reviewers at a machine-learning conference workshop in April 2025. This acceptance marks a significant milestone, described by its creators as the first time an autonomous system passed a "Turing test" for an AI-generated paper. While a "remarkable technological achievement," co-founder David Ha emphasizes that AI's role should be to aid human scientists rather than replace them, despite the system's ability to perform the full scientific discovery cycle.

Key takeaway

For AI Scientists evaluating autonomous research tools, AI Scientist's successful peer-reviewed publication and acceptance of an AI-generated paper at a conference workshop indicate a significant advancement in AI's research capabilities. You should consider integrating such tools as "co-scientists" to augment your research workflow, focusing on areas like hypothesis generation and experimental design, while acknowledging their current limitations compared to top human-produced work.

Key insights

AI Scientist, an autonomous research tool, successfully generated a peer-reviewed paper, demonstrating AI's capacity for scientific discovery.

Principles

• AI can automate the full scientific process.
• AI tools can pass human peer review.
• AI should augment human scientists.

Method

AI Scientist uses LLM agents (e.g., GPT-4o) to search literature, generate hypotheses, design research, write/execute code, and author papers, with an automated reviewer for quality control.

In practice

• Use AI agents for literature review.
• Employ LLMs for hypothesis generation.
• Integrate AI for automated code execution.

Topics

• AI Scientist
• Autonomous Research
• Large Language Models
• Peer Review
• Scientific Automation

Best for: AI Scientist, AI Researcher, Machine Learning Engineer, Research Scientist

Related on AIssential

• Towards the realization of AI research by AI: AI Scientist paper published in Nature journal — AI Scientist autonomously generates research papers, passes peer review, and reveals a scaling law for scientific quality.
• The AI Scientist: Towards Fully Automated AI Research, Now Published in Nature — AI agents can now autonomously conduct machine learning research, generate papers, and pass human peer review.
• AI-assisted Scientific Discovery, Experimentation, Content Generation, and Evaluation — AI tools are increasingly enhancing the scientific research lifecycle from discovery to peer review.
• Improving the academic workflow: Introducing two AI agents for better figures and peer review — AI agents can automate complex academic tasks like figure generation and peer review, improving efficiency and quality.
• How AI is changing the nature of mathematical research — AI tools are revolutionizing mathematical research by automating proof generation and contributing novel ideas, accelerating discovery.
• Why AI can’t be trusted to write scientific reviews — Current AI tools are not ready for high-stakes scientific systematic reviews due to limitations in context, nuance, and transparency.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine learning : nature.com subject feeds.