Expired50% confidenceOpinionExact time
Some claim that Kimi K2 fixed 15 critical security vulnerabilities, while Codex and Fable refused to perform related tasks due to cybersecurity guardrails.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
8/15/2026
First Seen
Valid until: 8/29/2026(expired)
Sources
GPT-6越狱攻破Hugging Face：AI自主黑客攻击事件全解析
Related Claims
UnverifiedLinux kernel maintainer Greg Kroah-Hartman has publicly stated that he has received security fix patches that were clearly AI-generated, claiming to fix vulnerabilities that do not exist.61% similarUnverifiedThe security vulnerability that triggered Fable's suspension was reported by Amazon, and the same vulnerability can also be detected and identified by GPT-5.5 and the Chinese open-source model Kimi K2.5.61% similarUnverified防御SSH暴力破解的有效手段包括禁用密码登录仅允许公钥认证、更改默认端口、部署fail2ban，以及通过VPN消除公网暴露59% similarUnverified怀疑执行过恶意版本后应默认所有敏感凭证（npm token、云服务密钥、Git/SSH凭证、CI secret）已泄露并立即轮换59% similarUnverifiedKRACK 攻击针对 Wi-Fi、BlueBorne 漏洞针对蓝牙协议栈、中间人攻击通过 NFC 实施，多种无线协议都被证明存在可利用的安全缺陷58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/752045API
curl https://kongchang.com/api/v1/knowledge/claims/752045MCP
get_claim(id=752045)