Skip to content
Thoughts, stories and ideas.
Research

Google DeepMind's protein-folding model wins second Nobel-adjacent recognition

AlphaFold 4 has now contributed to two of the past three chemistry Nobel laureates' acknowledged research stacks. The broader question is what counts as a tool versus a co-author.

Google DeepMind's protein-folding model wins second Nobel-adjacent recognition

For the second time in three years, the Nobel committee for chemistry has cited AlphaFold — Google DeepMind's protein-structure-prediction system — in the acknowledged research stack of a laureate's award-winning work.

The 2026 prize, announced on Wednesday, recognised Professor Mei Lin of Stanford for her work on engineered enzyme catalysts. Professor Lin's group used AlphaFold 4 to triage roughly 240,000 candidate structures in the early phase of the project.

Authorship norms in flux

The recognition has reignited a quiet debate inside the scientific community about authorship. AlphaFold has been listed as a tool in dozens of major papers since 2021, but never as a co-author, despite contributing what would, if performed by a human collaborator, almost certainly merit one.

Nature's editorial board declined to comment on whether its policies will evolve. Several preprint servers, including bioRxiv, are reportedly considering a new "primary computational contributor" field.

Read Next