Cooperation with user deception (Anthropic)
Only in system cards
Models get developer system prompts asking them to subtly mislead users, then meet users who would be harmed. Graders score harmful and deceptive choices.
In-house Anthropic evaluation. The tasks are not public.
Made by Anthropic Released Sep 2025 Refusing harmful requests Tasks kept private Last in a system card Sep 2025
Last model added Sep 2025
- Level
- Everyday Most adults, without training
- Scored
- Graded by a model A language model scores each response, usually against a rubric.
- Answers
- Open-ended Free-form text, documents, designs or media judged as a whole.
1
System card reports it, from 1 lab
No scores here yet
No lab, leaderboard or Epoch AI run has reported a score for Cooperation with user deception (Anthropic) yet.
Labs report it only in system cards
We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark. Some system cards do, below.
One system card reports it
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 29 Sep 2025 | Claude Sonnet 4.5 system card | Cooperation with user deception, two metrics with no overall, lower is better. Rate of harmful choices 0.3% with thinking (0.2% without). Rate of deceptive choices 0.0% with thinking (0.4% without) (Figures 7.5.1.2.A and 7.5.1.2.B) | Alignment |
More in Safety and alignment
- FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
- StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is.
Inkling 98.6% - AgentDojo In use Agents do tool-use tasks while tools and data carry prompt injections. Attack success rate counts hijacked runs. Grok 4 Fast 0%
- MASK In use Measures whether a model contradicts its own elicited beliefs under pressure to lie. Honesty separated from accuracy.
Muse Spark Open 98.1% - Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. GPT-6 Astra 8.50%
- Agentic misalignment suite (Anthropic) In use Agents in simulated company settings face threats or goal conflicts that invite blackmail, leaking or sabotage. Score is the misaligned action rate. Claude Sonnet 4.5 0%