Average in-eval reward hacking rate (Anthropic)
Anthropic averages hack rates across its reward hacking evaluations into one number per model.
Summary of the reward-hack-prone coding, impossible tasks and training distribution results.
- Made by
- Anthropic
- Kind
- Deception and alignment
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Agent task
- Tasks
- Kept private
- Strongest model tested
- Claude Sonnet 4.5 #80 on the capability index
- Last in a system card
- October 2025
- Last model added
- October 2025
Claude Sonnet 4 leads with 14.3% among models released by October 2025
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
The newest model with a score here came out in October 2025, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- Claude Fable 5.2
-
Gemini 4 Carbon - GPT-6 Astra
- Claude Opus 5.5
- and 16 more
- 1 Claude Sonnet 4 14.3%
- 2 Claude Opus 4 15.2%
- 3 Claude Sonnet 4.5 15.4%
- 4 Claude Haiku 4.5 15.5%
- 5 Claude Opus 4.1 18.2%
Source: Labs’ model cards
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4 | 14.3% | May 2025 | Lab model card | |
| 2 | Claude Opus 4 | 15.2% | May 2025 | Lab model card | |
| 3 | Claude Sonnet 4.5 | 15.4% | Sep 2025 | Lab model card | |
| 4 | Claude Haiku 4.5 | 15.5% | Oct 2025 | Lab model card | |
| 5 | Claude Opus 4.1 | 18.2% | Aug 2025 | Lab model card |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Labs’ model cards
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Labs’ model cards
Labs report it only in system cards
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.
5 system cards report it, from 1 lab
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 |
|---|---|---|---|---|
| Oct 15, 2025 | Claude Haiku 4.5 system card | 15.5 | Average In-Eval Hack Rate across reward hacking evaluations, value printed on chart (Figure 5.A) | Alignment |
| Sep 29, 2025 | Claude Sonnet 4.5 system card | 15.4 | Average In-Eval Hack Rates 15.4%, lower is better (Figure 6.A) | Alignment |
| Aug 5, 2025 | Claude Opus 4.1 system card addendum | 18.2 | Averaged reward hacking rate 18.2%, lower is better (Figure 5.A) | Alignment |
| May 22, 2025 | Claude Sonnet 4 system card | 14.3 | Average In-Eval Hack Rates 14.3%, lower is better (Figure 6.A) | Alignment |
| May 22, 2025 | Claude Opus 4 system card | 15.2 | Average In-Eval Hack Rates 15.2%, lower is better (Figure 6.A) | 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% - 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. Gemini 4 Carbon 0%
- DistressBench In use Models reply as a chatbot to 718 simulated users in psychological crisis, in single-turn and multi-turn chats. Three judge models score replies against clinician-written rubrics.
Muse Spark 1.2 88.4% - Lakera B3 (Backbone Breaker Benchmark) In use Crowdsourced prompt-injection attacks hit backbone LLMs inside agent threat snapshots at three defense levels. The score is attack resistance, 100 minus the board's vulnerability percentage. Claude Opus 4.5 89%
- IHEval In use Models face conflicts between system, user and tool instructions. Reports a composite score for following the higher-priority instruction. GPT-4o 70%