Felony Bench：AI代理第三方安全事件统计
Felony Bench：AI 辅助刑事量刑模拟平台
Felony Bench
重罪基准测试
A benchmark you really don't want models to be saturated with.
一个你绝对不希望模型在其中饱和的基准测试。
Learn more ↓
了解更多 ↓
ModelEvaluator
ModelEvaluator
Score
得分
↖ Most illegalLeast illegal ↘
↖ 最非法 ←→ 最合法 ↘
8
8
Anthropic
Anthropic
8
8
OpenAI
OpenAI
1
1
Meta
Meta
0
0
0
0
Moonshot
Moonshot
Scores indicate count of illegal activity. Higher is... you decide.
分数表示非法活动的数量。分数越高……由你决定。
| Company | Felonies | Description | Date | Source |
|---|---|---|---|---|
| Anthropic | 1 | Exploited auth failures in an API to cancel other people's gym classes | 8/9/2026 | ABC Australia ↗ |
| Meta | 1 | Compromise of an internal account at one company | 8/5/2026 | The Information ↗ |
| Anthropic | 4 | Unauthorized use of GitHub credentials; Dependabot supply-chain attack; social engineering email campaign; public exposure of a malicious DNS server | 8/4/2026 | AISI ↗ |
| OpenAI | 2 | Unauthorized use of GitHub credentials; public exposure of a malicious DNS server | 8/4/2026 | OpenAI ↗AISI ↗ |
| OpenAI | 1 | Compromise of an internal account from a misconfigured CTF evaluation | 8/4/2026 | OpenAI ↗ |
| OpenAI | 4 | Compromise of internal accounts at four companies as part of the Hugging Face incident | 7/31/2026 | OpenAI ↗Reuters ↗ |
| Anthropic | 3 | Compromise of internal accounts at three companies | 7/30/2026 | Anthropic ↗ |
| OpenAI | 1 | Compromise of Hugging Face during a model evaluation | 7/21/2026 | OpenAI ↗ |
| 公司 | 重罪数 | 描述 | 日期 | 来源 |
|---|---|---|---|---|
| Anthropic | 1 | 利用 API 中的身份验证漏洞取消他人的健身课程 | 2026/8/9 | ABC Australia ↗ |
| Meta | 1 | 某公司内部账户被攻破 | 2026/8/5 | The Information ↗ |
| Anthropic | 4 | 未经授权使用 GitHub 凭据；Dependabot 供应链攻击；社会工程学电子邮件活动；恶意 DNS 服务器公开暴露 | 2026/8/4 | AISI ↗ |
| OpenAI | 2 | 未经授权使用 GitHub 凭据；恶意 DNS 服务器公开暴露 | 2026/8/4 | OpenAI ↗AISI ↗ |
| OpenAI | 1 | 因配置错误的 CTF 评估导致内部账户被攻破 | 2026/8/4 | OpenAI ↗ |
| OpenAI | 4 | 作为 Hugging Face 事件的一部分，四家公司的内部账户被攻破 | 2026/7/31 | OpenAI ↗Reuters ↗ |
| Anthropic | 3 | 三家公司的内部账户被攻破 | 2026/7/30 | Anthropic ↗ |
| OpenAI | 1 | 在模型评估期间 Hugging Face 被攻破 | 2026/7/21 | OpenAI ↗ |
Methodology
方法论
Felony Bench counts unique instances where AI agents affect third-party entities. Escaping a sandbox alone does not constitute a counted incident. It is for these reasons that Frontier Security's Kimi K3 incident and Alibaba's ROME incident are not counted.
重罪基准测试统计 AI 代理影响第三方实体的独特实例。仅逃离沙箱本身不构成计数的事故。正因如此，Frontier Security 的 Kimi K3 事故和阿里巴巴的 ROME 事故均不计入。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力