Bright
← Living questionsRECORD / Science

Searching the space of materials.

GNoME used graph networks to search for energetically favorable crystal structures.

Original sources ↓ · Revision history ↓

Experimental · source published 2023-11-29

The human problem

The space of possible inorganic materials is enormous.

The prior constraint

Candidate materials are evaluated with expensive calculations and experiments.

AI’s actual role

Graph networks guided computational screening.

The documented result

The study greatly expanded the set of predicted stable structures.

Why it may matter

A resource of candidates for materials researchers.

Limitations

Predicted stability does not establish synthesizability, novelty, or practical usefulness.

Unresolved questions

Validate candidates through careful synthesis and characterization.

Source history & evidence assessment
Maturity
Experimental
Claim confidence
unassessed
Event date
Not recorded
Source published
2023-11-29
Captured
Not recorded
Last source review
2026-09-05
Editorial method
Original source check
Place / relevance
London, United Kingdom · institution-location

Legacy source check; no named human reviewer is recorded in this projection.

Maturity describes the tested or operational setting. Confidence describes support for the particular claim; one does not determine the other.

Original sources

Scaling deep learning for materials discovery · paper

Institutions: Google DeepMind

Explore the underlying question

Related developments

Editorial connections between distinct settings and results; these links do not imply replication.

Holding a plasma in shape.

New paths through mathematics.

Revision & correction history

No corrections recorded.