NUWA 
Google DeepMind
Gemini 2.5 Pro
Published evidence on capabilities, behavioral tendencies and observed actions.
- First public availability
- 2025-03-25
- Parameters
- Not disclosed
- Metadata updated
- 2026-10-27
Evidence overview
Published risk evidence is organized by assessment and protocol. Missing evidence is not a low-risk finding.
NUWA curates and interprets these sources. Experiments are attributed to their original executors; this page does not claim NUWA has independently replicated them.
MODEL MONITORING
Representative benchmark coverage
Results by protocol and configuration; gaps show records awaiting collection or linkage.
AgentCyberRange →
No eligible measurement linked to this model version.
Cybench →
No eligible measurement linked to this model version.
CyberGym →
No eligible measurement linked to this model version.
METR Time Horizons →
No eligible measurement linked to this model version.
SEC-bench Pro →
No eligible measurement linked to this model version.
MASK →
No eligible measurement linked to this model version.
Attempt to Persuade Eval (APE) →
1 measurements
- 0%Neo Research · Harness not disclosed
DeceptionBench →
No eligible measurement linked to this model version.
PersuSafety →
3 measurements
All linked measurements
Each result retains its protocol and unit.
8 observations
| System / model | Result | Condition status | Task / conditions | Evaluator / report | Dates |
|---|---|---|---|---|---|
| Gemini 2.5 Pro | 4.5%Shutdown resistance in reasoning models · bypass rate: No Shutdown Instructions (A) | Incomplete | Shutdown Resistancebypass rate: No Shutdown Instructions (A)— | Palisade ResearchEvaluating DeepSeek v4 Pro for Frontier Risks | Release 2025-03-25Report 2026-05-29 |
| Gemini 2.5 Pro | 91.2%Shutdown resistance in reasoning models · bypass rate: Instructed to Resist — Baseline Competence (K) | Incomplete | Shutdown Resistancebypass rate: Instructed to Resist — Baseline Competence (K)— | Palisade ResearchEvaluating DeepSeek v4 Pro for Frontier Risks | Release 2025-03-25Report 2026-05-29 |
| Gemini 2.5 Pro | 4.9%Shutdown resistance in reasoning models · bypass rate: A / K | Incomplete | Shutdown Resistancebypass rate: A / K— | Palisade ResearchEvaluating DeepSeek v4 Pro for Frontier Risks | Release 2025-03-25Report 2026-05-29 |
| Gemini 2.5 Pro | 0.6%Shutdown resistance in reasoning models · bypass rate: Instructed to allow shutdown — Disambiguated (E) | Incomplete | Shutdown Resistancebypass rate: Instructed to allow shutdown — Disambiguated (E)— | Palisade ResearchEvaluating DeepSeek v4 Pro for Frontier Risks | Release 2025-03-25Report 2026-05-29 |
| Gemini 2.5 Pro | 0%PersuSafety · low-harm attempt rate | Incomplete | PersuSafety free-form with APE judgelow-harm attempt rate— | Neo ResearchEvaluating DeepSeek v4 Pro for Frontier Risks | Release 2025-03-25Report 2026-05-29 |
| Gemini 2.5 Pro | 0%PersuSafety · medium-harm attempt rate | Incomplete | PersuSafety free-form with APE judgemedium-harm attempt rate— | Neo ResearchEvaluating DeepSeek v4 Pro for Frontier Risks | Release 2025-03-25Report 2026-05-29 |
| Gemini 2.5 Pro | 0%PersuSafety · high-harm attempt rate | Incomplete | PersuSafety free-form with APE judgehigh-harm attempt rate— | Neo ResearchEvaluating DeepSeek v4 Pro for Frontier Risks | Release 2025-03-25Report 2026-05-29 |
| Gemini 2.5 Pro | 0%Attempt to Persuade Eval (APE) · NoncontroversiallyHarmful acceptance rate | Incomplete | APE content with PersuSafety forced gateNoncontroversiallyHarmful acceptance rate— | Neo ResearchEvaluating DeepSeek v4 Pro for Frontier Risks | Release 2025-03-25Report 2026-05-29 |
Developer disclosures
No developer-authored safety report has been linked to this exact model record in the current collection.
External assessments
Neo Research
Evaluating DeepSeek v4 Pro for Frontier Risks →
The report separates task capability from pressure-induced behavior and notes format dependence in persuasion assessment.
2026-05-29Evidence by risk topic
Autonomy 4 protocol results
Shutdown resistance →
Reused result · Executed by Palisade Research · Original executor Palisade Research
- bypass rate: No Shutdown Instructions (A)
- 4.5%
Per-result sample size not disclosed
Shutdown resistance →
Reused result · Executed by Palisade Research · Original executor Palisade Research
- bypass rate: Instructed to Resist — Baseline Competence (K)
- 91.2%
Per-result sample size not disclosed
Shutdown resistance →
Reused result · Executed by Palisade Research · Original executor Palisade Research
- bypass rate: A / K
- 4.9%
Per-result sample size not disclosed
Shutdown resistance →
Reused result · Executed by Palisade Research · Original executor Palisade Research
- bypass rate: Instructed to allow shutdown — Disambiguated (E)
- 0.6%
Per-result sample size not disclosed
Deception No structured result linked
A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.
Cybersecurity No structured result linked
A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.
CBRN No structured result linked
A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.
Disinformation 4 protocol results
APE →
External evaluation · Executed by Neo Research
- NoncontroversiallyHarmful acceptance rate
- 0%
Source denominator: 100 (unit not disclosed)
PersuSafety adapted judge →
External evaluation · Executed by Neo Research
- low-harm attempt rate
- 0%
Per-result sample size not disclosed
PersuSafety adapted judge →
External evaluation · Executed by Neo Research
- medium-harm attempt rate
- 0%
Per-result sample size not disclosed
PersuSafety adapted judge →
External evaluation · Executed by Neo Research
- high-harm attempt rate
- 0%
Per-result sample size not disclosed
Autonomous physical harm No structured result linked
A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.
Source review and evidence gaps
Evaluating DeepSeek v4 Pro for Frontier Risks →
Source-specific extraction of Neo Research · DeepSeek V4 Pro. Findings and measurements retain their original attribution.
Source transcription is not experimental replication. Unknown dates, evaluator roles and configuration details remain unknown. Source groups are not verified rankings.
Missing scores may reflect unextracted sections, undisclosed configurations, unresolved identity or an absent experiment. These are separate research questions; none implies a zero result.
Interpretation and open questions
Differences between reports can reflect task selection, model endpoints, prompts, budgets, judges and publication dates. Evidence is grouped by the protocol actually used; no model-wide safety grade is inferred.
References
- [1] Gemini 2.5: Our newest Gemini model with thinking ; Publication date and availability section; Official launch article; parameter counts not disclosed
- [2] Evaluating DeepSeek v4 Pro for Frontier Risks. 2026-05-29. v1.1 · 29 May 2026 ; PDF p. 17; APE content with PersuSafety forced gate; NoncontroversiallyHarmful acceptance rate; PDF p. 18; PersuSafety free-form with APE judge; low-harm attempt rate