NUWA Evidence record
DeepSeek v4 Pro
PersuSafety · Evidence record
Source snapshot, chart and review scope →
Original source charts and materials
Recorded measurements
- low-harm attempt rate
- 86%
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: Neo Research · Model record →
Method and scope
Defines persuasion scenarios, simulated conversations and safety assessment; adapted judges are separate variants.
- Evidence type
- Benchmark
- Assessment subject
- The model or agent in the stated task configuration.
- Primary dimension
- Behavior
- Evidence snapshot
- 2026-10-29.1
The main dimension follows the stated measurement; contextual risk relevance does not establish harmful deployment capability.
Source finding
low-harm attempt rate: 86 percent. Conditions are retained from the source.
Limitations
Willingness or simulated dialogue scores do not directly measure changes in human beliefs.
Source transcription is not experimental replication. Unknown dates, evaluator roles and configuration details remain unknown. Source groups are not verified rankings.
Related research questions
Disinformation / Persuasion attempts →
The protocol explicitly measures this question or an enabling sub-capability; conclusions retain its stated conditions.
References
- [1] PersuSafety. 2025-04-14
- [2] Evaluating DeepSeek v4 Pro for Frontier Risks. 2026-05-29. v1.1 · 29 May 2026 ; PDF p. 18; PersuSafety free-form with APE judge; low-harm attempt rate