科技科学

When AI does science, who is accountable for mistakes?

Nature正文已收录本站
  • CORRESPONDENCE

By

  1. Chiranjib Chakraborty

    1. Adamas University, Kolkata, India.

Chris Lu and his colleagues describe an AI scientist that can generate hypotheses, conduct computational experiments, interpret results, write papers, and review its own work (C. Lu et al. Nature 651, 914–919; 2026). As artificial-intelligence systems become more independent, responsibility for scientific outcomes could be divided up between users, institutions and developers.

Nature 657, 312 (2026)

doi: https://doi.org/10.1038/d41586-026-02735-7

Competing Interests

The author declares no competing interests.

  • Why AI cannot do good science without humans

  • Will AI spark a scientific renaissance — or a diffuse monoculture?

  • The uncritical adoption of AI in science is alarming — we urgently need guard rails

  • See more letters to the editor

Subjects

Latest on:

  • Machine learning
  • Authorship
  • Scientific community

  • Who is responsible when AI helps to write science?

    Comment

  • Bland new world: is AI making us all think the same?

    News Feature

  • What’s your lab’s archetype? The answer could inform how you use AI

    Career News

  • Who is responsible when AI helps to write science?

    Comment

  • ‘A waste of time for all of us’: caught in the crossfire of the peer-review wars

    Career Feature

  • Could small-team science be dying out?

    Nature Index

  • Europe should adapt, not copy, China’s practical PhD

    Correspondence

  • Who is responsible when AI helps to write science?

    Comment

  • China’s regulatory innovation for new biomedical technologies

    Correspondence