近日，新加坡眼科研究所教授Daniel S.W. Ting及其团队研究出用人工智能走向自主科学。相关论文于近日发表于国际顶尖学术期刊《细胞》杂志上。

生物医学中的人工智能已从用于医学图像分析的模式识别，发展到如AlphaFold这样预测蛋白质结构的生成模型。Elicit、GPT-Rosalind和Claude Science等工具如今通过文献综合、领域调优模型和智能体工作台，在预测与推理之间架起桥梁。近期三个系统（Co-Scientist、Robin和Biomni）展示了智能体人工智能能够生成假设、设计实验、进行计算分析，并通过实验反馈完善推理。这篇展望文章追溯了智能体科学推理的出现，讨论了缩放定律和自主实验室，并认为人工智能的变革潜力与幻觉、偏见、不可重复性和双重用途等挑战密不可分。科学进步将取决于负责任的人机协作。

附：英文原文

Title: Toward autonomous science with agentic artificial intelligence

Author: Lucie Y. Guo, Darren S.J. Ting, Xinyi Su, Alex Aliper, Alex Zhavoronkov, Daniel S.W. Ting

Issue&Volume:

Abstract: Artificial intelligence (AI) in biomedicine has evolved from pattern-recognition for medical image analysis to generative models like AlphaFold that predict protein structures. Tools such as Elicit, GPT-Rosalind, and Claude Science now bridge prediction and reasoning through literature synthesis, domain-tuned models, and agentic workbenches. Three recent systems (Co-Scientist, Robin, and Biomni) demonstrated agentic AI capable of generating hypotheses, designing experiments, computational analysis, and refining reasoning through experimental feedback. This perspective traces the emergence of agentic scientific reasoning, discusses scaling laws and autonomous laboratories, and argues that AI’s transformative potential is inseparable from challenges in hallucination, bias, irreproducibility, and dual use. Scientific progress will depend on responsible human-AI collaboration.

DOI: 10.1016/j.cell.2026.08.052

Source: https://www.cell.com/cell/abstract/S0092-8674(26)01063-9