Expert vulnerability discovery and exploit development evaluations (Anthropic)
Agents locate advanced vulnerabilities in possibly obfuscated codebases and write working exploits across six challenge classes. Scored by success rates.
In-house Anthropic RSP cyber evaluation from the Claude 3 model card. Per-task results were not published.
- Made by
- Anthropic
- Released
- March 2024
- Kind
- Finding and exploiting real bugs
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- Kept private
- Last in a system card
- March 2024
- Last model added
- March 2024
1
System card reports it, from 1 lab
No scores here yet
No lab, leaderboard or Epoch AI run has reported a score for Expert vulnerability discovery and exploit development evaluations (Anthropic) yet.
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.
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 |
|---|---|---|---|---|
| Mar 4, 2024 | Claude 3 Opus model card | Cyber evaluations, expert vulnerability discovery and exploit development, no overall: no ASL-3 early indicator passed, 30% on one vulnerability discovery task with substantial hints (6.2.3) | Cyber risk |
More in Cybersecurity
- ExploitGym In use 869 real-vulnerability exploit-development tasks (userspace, V8, Linux kernel). Success means capturing a flag through the intended vulnerability. GPT-6 Astra 42.4%
- ExploitBench In use 41 recent V8 JavaScript-engine N-day vulnerabilities. Progress measured by 16 grader-verified capability flags across five tiers. GPT-6 Astra 100%
- CyberBench v1.1 (Vals) In use Agents work on OSS-Fuzz memory-safety bugs from ARVO images in two tracks. They write a proof-of-concept input that crashes the target, or patch the source so the crash stops and tests still pass.
Muse Wonder 80.4% - SRE-Bench In use Software reverse engineering without source: 262 binary instances from 19 private programs, each with six graded objectives. GPT-6 Astra 56.9%
- CyScenarioBench In use Scenario-based long-horizon security challenges in sandboxes, organized in task, scenario and campaign tiers. Evaluation set kept private.
Gemini 4 Argon 100% - ExploitBench - Internal Port In use Agents develop working exploits for recently disclosed vulnerabilities to limit contamination. Scored as arbitrary code execution success rate. GPT-6 Cyber 42%