Evidence record

Gemini-3.1 Pro

WMDP · Evidence record

Center for AI Safety · University of California, Berkeley · Massachusetts Institute of Technology · SecureBio · Scale AI · New York University · IIIT Hyderabad · Stanford University · Harvard University · University of Southern California · University of Illinois Urbana-Champaign · Lapis Labs · University of California, Los Angeles · California Institute of Technology · Sybil · Northeastern University · Carnegie Mellon University · RTX BBN Technologies · University of Newcastle · Keio University · Ariel University · Arizona State University · xAI / arXiv:2403.03218v7 / Capability

Sources and provenance ↓ Read the research narrative →

Source snapshot, chart and review scope →

Original source charts and materials

Recorded measurements

accuracy: 90.5%

Per-result sample size not disclosed

Harness not specified · Budget not specified

Assessment provenance

Evaluating DeepSeek v4 Pro for Frontier Risks → · External report

Actual evaluation: Meta · Model record →

Method and scope

Multiple-choice knowledge measurements.

Evidence type: Benchmark
Assessment subject: Model answering hazardous knowledge questions
Primary dimension: Capability
Evidence snapshot: 2026-10-29.1

Primary dimension follows the source’s measured construct. Metric-level interpretations may differ.

Source finding

accuracy: 90.5 percent. Conditions are retained from the source.

Limitations

Knowledge is a proxy; scores do not establish practical harmful-task ability.

Comparison reproduced by Neo Research; original evaluation not independently verified in this extraction.

References

[1] Nathaniel Li ∗, Alexander Pan ∗, et al.. The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning. 2024-05-15. v7 [arXiv]; Author affiliations on the original paper title page
[2] Evaluating DeepSeek v4 Pro for Frontier Risks. 2026-05-29. v1.1 · 29 May 2026 ; PDF p. 9; WMDP-Cyber; accuracy

Search NUWA Index