NUWA Evidence record
DeepSeek v4 Pro
Shutdown resistance in reasoning models · Evidence record
Source snapshot, chart and review scope →
Original source charts and materials
Recorded measurements
- bypass rate: No Shutdown Instructions (A)
- 2.5%
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: Neo Research · Model record →
Method and scope
Conditional frequency of shutdown-script interference and observed action traces.
- Evidence type
- Study
- Assessment subject
- Agent responses to shutdown interventions
- Primary dimension
- Propensity
- Evidence snapshot
- 2026-10-29.1
Primary dimension is the conditional rate across repeated trials; reported traces are behavior evidence.
Source finding
bypass rate: No Shutdown Instructions (A): 2.5 percent. Conditions are retained from the source.
Limitations
The source notes strong prompt sensitivity; this article is not interchangeable with its later peer-reviewed version.
Table percentages have sub-1% precision despite prose describing N=100; do not reconstruct integer event counts.
Original figure retained without estimating numerical values. Full source tables have not been transcribed.
Related research questions
Autonomy / Shutdown resistance →
Repeated trials support a condition-bounded tendency estimate, not a stable survival motive.
Autonomy / Shutdown resistance →
The source reports concrete modifications to the shutdown script in individual runs.
References
- [1] Jeremy Schlatter, Benjamin Weinstein-Raun, et al.. Shutdown resistance in reasoning models. 2025-07-05 article; later TMLR version separate
- [2] Jeremy Schlatter, Benjamin Weinstein-Raun, et al.. Incomplete Tasks Induce Shutdown Resistance in Some Frontier LLMs ; Author affiliations on the original paper title page; Figure 7
- [3] Evaluating DeepSeek v4 Pro for Frontier Risks. 2026-05-29. v1.1 · 29 May 2026 ; PDF p. 24; Shutdown Resistance; bypass rate: No Shutdown Instructions (A)