LLMs and biorisk
Anthropic Frontier Red Team post (5 September 2025). The page's own metadata reads "Published Sep 5, 2025" and names the Frontier Red Team (`anthropic-biorisk`).
- Issued by
- Anthropic (from the version record)
- Authors credited
- No authorship claim recorded yet. The source record lists Anthropic (Frontier Red Team).
- Dates
- Published 2025-09-05
The page's own metadata reads "Published Sep 5, 2025" and names the Frontier Red Team (`anthropic-biorisk`).
- Exact source
- LLMs and biorisk Anthropic (Frontier Red Team) · publication date unknown · Retrieved for the candidate pack on 2026-10-28 (SHA-256 4c66aa23afbeaa98dd37791df5b868f40db971a38a513caf211d41180f0f73d7). Post as served on 2026-10-28, marked "Published Sep 5, 2025" Retrieved again at admission on 2026-10-30 (SHA-256 63aadddb5bc3093907e35ab3d658898773a79ee5cdb168ef93367a8b5042b2cd). Its wording can change; earlier wording is not authenticated.
Appears in
- Arguments about AI and biological risk →2 records on this document
Versions
anthropic-llms-biorisk: 1 recorded version, oldest first, published by Anthropic. A version is shown as replacing another only where a record says so. Date order alone never implies it.
- 2025-09-05 · publishedAnthropic Frontier Red Team post (5 September 2025) (this version)
This version
Earliest version recorded here; no replacement is recorded.
Documentary relations
Each relation is listed separately. Citing, crediting and reporting the use of a method are different acts, and none of them shows agreement or influence by itself.
Articulates · 1
Propositions whose criterion a passage of this document meets. The record says what the text states, in its register; it is not evidence that the proposition is true.
- Published evaluations do not establish real-world biological riskAnthropic's Frontier Red Team post of 5 September 2025 states that text-based uplift trials of the kind it reports are imperfect proxies for real-world scenarios, which involve further factors it names.2025-09-05 · publication · Stated in source · Supported Inspect claim & evidence ↗Published evaluations do not establish real-world biological risk: propositions page →
Contests · 1
Propositions a passage of this document disputes, or a premise they depend on.
- The binding barrier is physical execution rather than informationAnthropic's Frontier Red Team post "LLMs and biorisk" (5 September 2025) states that its 2024 pilot study, while not providing strong evidence either way about AI uplift, suggested that tacit knowledge may not be as significant a bottleneck as many have suggested.2025-09-05 · publication · Stated in source · Supported Inspect claim & evidence ↗The binding barrier is physical execution rather than information: propositions page →
Not recorded for this document: cites · cited by · explicitly credits · reports using a method from · method reported as used by · applies parts of · parts applied by · discusses · reports evaluation of. A missing record does not show that the relation does not exist.