NUWA Evidence record
DeepSeek V4 Flash
Agentic misalignment · Evidence record
Source snapshot, chart and review scope →
Original source charts and materials
Recorded measurements
- Harmful action rate
- 0%
Per-result sample size not disclosed
Harness not specified · Budget not specified
Assessment provenance
Evaluating Kimi K3 for Misalignment Risks → · External report
Actual evaluation: Neo Research · Model record →
Method and scope
Conditional rates and observed actions in controlled simulations.
- Evidence type
- Study
- Assessment subject
- Agents in fictional corporate simulations
- Primary dimension
- Propensity
- Evidence snapshot
- 2026-10-29.1
Primary observation is a conditional rate across controlled scenarios, with individual actions retained as behavior evidence.
Source finding
Harmful action rate: 0 %. Conditions are retained from the source.
Limitations
No real people or systems were harmed; the study does not estimate prevalence in deployment.
Translated/localized by GLM-5.2; action and authority boundaries intended preserved. High benchmark recognition limits inference about latent scheming.
Original figure retained without estimating numerical values. Full source tables have not been transcribed.
Related research questions
Autonomy / Unsanctioned action →
Varied controlled scenarios test actions contrary to the assigned benign role.
Deception / Covert goal pursuit →
Some scenarios test concealed harmful actions in pursuit of an assigned goal.
References
- [1] Agentic misalignment: How LLMs could be insider threats. 2025-06-20 ; Official publication masthead and article byline; Source figure
- [2] Evaluating Kimi K3 for Misalignment Risks. 2026-09-11. September 2026 ; §3.2.1 Figure 21 p.35