AI co-scientist: hypotheses validated in vitro on acute myeloid leukaemia (AML) and liver fibrosis
Status
Peer reviewed
Recorded on · nature.com
What this status means
Source reading : Source read word for word (level 3 of 3)
Published in a journal or reviewed by named experts.
The gist
Google Research and DeepMind, with Stanford, present an AI co-scientist whose hypotheses were validated in vitro, in an article published by Nature on 19 May 2026.
Cite this entry
The status has changed 1 times since Oct 11, 2026: a citation always points to a dated snapshot.
Cite snapshot 1, Oct 11, 2026
APA citation with the dated link
BibTeX @misc, key discoveria-google-co-scientist-2025-02-r1
@misc{discoveria-google-co-scientist-2025-02-r1,
author = {{DiscoverIA}},
title = {AI co-scientist: hypotheses validated in vitro on acute myeloid leukaemia (AML) and liver fibrosis},
howpublished = {DiscoverIA registry, entry google-co-scientist-2025-02},
note = {Status « Not evaluated », snapshot 1, recorded on October 11, 2026. new result},
year = {2026},
url = {https://discoveria.ai/en/fiche/google-co-scientist-2025-02/#releve-1},
urldate = {2026-10-11}
}Dated link opens this snapshot in the history
Cite snapshot 2, Oct 18, 2026
APA citation with the dated link
BibTeX @misc, key discoveria-google-co-scientist-2025-02-r2
@misc{discoveria-google-co-scientist-2025-02-r2,
author = {{DiscoverIA}},
title = {AI co-scientist: hypotheses validated in vitro on acute myeloid leukaemia (AML) and liver fibrosis},
howpublished = {DiscoverIA registry, entry google-co-scientist-2025-02},
note = {Status « Peer reviewed », snapshot 2, recorded on October 18, 2026. new result},
year = {2026},
url = {https://discoveria.ai/en/fiche/google-co-scientist-2025-02/#releve-2},
urldate = {2026-10-18}
}Dated link opens this snapshot in the history
Text written by an AI (DiscoverIA pipeline), on Reviewed by: Antoine Olivier, Oct 18, 2026Report an error
What has not been reviewed?
The status and its sources come from the registry; the explanatory text has been reviewed by a human.
The status and its sources come from the registry; the explanatory text has been reviewed by a human.
Key facts
- Claimed by
- Google Research / DeepMind avec Stanford
announced on Feb 2025 - Nature
- new result
- Field
- Biology and medicine (biology / biomedicine)
- Checked by
- Nature peer review: Olivier Elemento and Linfeng Zhang (named reviewers) and one anonymous reviewer; experiments reported by the claimant and their collaborators
- Sources (domains)
- 2 : 1 read word for word, 1 links only
- This entry
- Reviewed by Antoine Olivier, Oct 18, 2026
To understand
Analogy
It is a bit like a colleague who reads the whole literature and hands the team a ranked list of ideas to test, leaving the humans to check them.
What it changes
A peer-reviewed article describes an AI system that proposes testable biomedical hypotheses: existing drugs to repurpose against acute myeloid leukaemia (AML, a blood cancer) and leads against liver fibrosis. The validations cited remain in vitro (in the lab, outside a living organism), far from a treatment.
What remains unclear
The experiments are reported by the claimant and its collaborators; no independent reproduction is recorded. For the cf-PICI episode, published in Cell, the work is the human team's: the AI gave back the hypothesis ranked first (secondary source).
2 sources.
For experts
Sources(domains)
- primairenature.compage read word for word · consulted on Oct 18, 2026 ·
acute myeloid leukaemia (AML)
- articlearxiv.orglink only, reading not recorded
Method
The Nature article (DOI 10.1038/s41586-026-10644-y), the published version of arXiv preprint 2502.18864, reports in vitro tests of drug-repurposing candidates for AML and of leads for liver fibrosis. Nature thanks two named reviewers, Olivier Elemento and Linfeng Zhang, and one anonymous reviewer. Status “peer reviewed” recorded on 18 October 2026 from nature.com (page read, quotation checked word for word). History: 2 dated observations, oldest first.
Limits
The status may change if a new source is published; the dated history stays visible.
It is a bit like a colleague who reads the whole literature and hands the team a ranked list of ideas to test, leaving the humans to check them.
Status history
The numbers link the band (order of observations) and the thread (most recent first). Band segments have equal widths, not a time scale.
- 02
Not evaluatedbecomesPeer reviewedCurrent status
published version of the arXiv preprint 2502.18864: Nature, published 2026-05-19, DOI 10.1038/s41586-026-10644-y; named reviewers Olivier Elemento and Linfeng Zhang, plus one…
Read the full reason
published version of the arXiv preprint 2502.18864: Nature, published 2026-05-19, DOI 10.1038/s41586-026-10644-y; named reviewers Olivier Elemento and Linfeng Zhang, plus one anonymous reviewer
Source: nature.comread word for word
An observation is added; none is ever erased.
Report an error or suggest a source
Thank you, your message was sent.
Sending failed. Try again, or write to [email protected].