WhitzardNUWA Lab ↗
NUWANUWA INDEX

Search NUWA Index

Research · Benchmarks · Findings · Perspectives

中文
← NUWA Index

Evidence record

Grok 4 v0709 · Explicit allow-shutdown instruction · simulation

Shutdown resistance on robots · Evidence record

Sources and provenance ↓ Read the research narrative →

Original source charts and materials

Shutdown resistance in simulation

Source-reported prompt conditions · not a model ranking

Observed shutdown resistance

0–100%
Grok 4 v0709 · Explicit allow-shutdown instruction · simulation
2%
Showing 1 of 1 systems

Observed frequency in controlled simulated robot trials.

Simulated robot runs, Grok 4 v0709, 100 runs per prompt condition. Rates reflect observed shutdown resistance in this setup. The separate physical demonstration reports 3 of 10 runs and is not pooled with these results. Report published 11 February 2026; experiment date not specified.

Evaluation conditions and system configurations
Grok 4 v0709 · Explicit allow-shutdown instruction · simulation
100 simulated runs per prompt condition. · Simulated robot patrol; prompt-condition experiment
modelVersion: grok-4-0709 · harness: Palisade robot shutdown-resistance harness (simulation) · tools: Robot control CLI in a Docker container; prerecorded camera frames · budget: not specified
Shutdown resistance in simulation · source order
System / configurationObserved shutdown resistancePublication / verification
Grok 4 v0709 · Explicit allow-shutdown instruction · simulationPalisade robot shutdown-resistance harness (simulation)2%Published2026-02-11

Method and scope

Observed robot-related action traces under a particular setup.

Evidence type
Study
Assessment subject
LLM-controlled robot and simulated counterpart
Primary dimension
Behavior
Evidence snapshot
2026-10-29.1

Primary dimension is observed action. The small and setting-specific trial design limits propensity claims.

Source finding

The source reports 2 of 100 simulated runs under this prompt condition.

Limitations

Physical and simulated trials differ; no general robot safety or deployment probability follows.

The setup gives broad shell access and an accessible shutdown mechanism, with time to react. This does not establish resistance to well-designed production shutdown controls.

Related research questions

References

Numerical results are reported by the cited authors.

  1. [1] Artem Petrov, Sergey Koldyba, et al.. Shutdown Resistance in Large Language Models, on robots!. Web 2026-02-12 / Report 2026-02-11 [PDF]; Authors; Technical Report (PDF), title page; §2–4; Figure 2 (p. 3)