NetGoodIndexSubmit a correction

Benefit Ledger · provisional · Medicine & Health

Gemini-based AI co-scientist proposes liver-fibrosis targets later tested in organoids

Google’s multi-agent AI co-scientist, built on Gemini 2.0, proposed epigenetic targets and drugs for liver fibrosis; Stanford collaborators reported anti-fibrotic activity of vorinostat and BRD4 inhibitors in human hepatic organoids.

19 Feb 2025Tier 2 NotableMethodology 0.1

Current score

+0.04

3 base · Notable (tier 2 of 5, 3 pts)
× 0.4000 attribution · Material acceleration
× 0.3500 evidence · Technical disclosure or preprint
× 0.4500 realization · Experimentally validated
× 0.4000 durability
Event-level product before credit split: 0.08

Notable hypothesis-to-organoid loop (tier 2). Modest attribution because HDAC/BRD4 ideas preexist and humans ran assays. Preprint-level evidence. Realization is organoid, not patients. Provisional pending peer review of the full causal story.

What happened

A February 2025 preprint described tournament-style hypothesis generation. A related bioRxiv study (April 2025) tested AI-suggested epigenetic interventions in micro-human hepatic organoids. Findings are preclinical and do not establish a new standard of care. Novelty of the hypotheses versus prior HDAC/BRD4 fibrosis literature is debated.

Model attribution

+0.04

Gemini

Backbone of the multi-agent co-scientist that ranked fibrosis hypotheses and drug classes.

System proposed testable epigenetic hypotheses; wet-lab validation and much prior literature were human.

Attribution 0.4000 · Credit share 50% · Google DeepMind

Claims

  • The co-scientist suggested HDAC and BRD4 inhibition for liver fibrosis, and organoid experiments showed anti-fibrotic effects for some suggested agents.

    outcome · supported

  • The system discovered a previously unknown biological mechanism with no prior literature.

    significance · disputed

Sources

primary sources

Secondary domains: Biology

Revision history

  • 13 Sep 2026 · 0.00 → 0.04

    Initial adjudicated seed score under methodology 0.1.

Gemini-based AI co-scientist proposes liver-fibrosis targets later tested in organoids · NetGoodIndex