Unverified50% confidenceFactExact time
The BioBaseExpand benchmark test data comes from an interactive chart seemingly leaked on OpenAI's website, has not yet been independently verified, and the related issue is said to have been fixed.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
11/18/2026
First Seen
Valid until: 12/2/2026
Sources
GPT-6.2 Astra Leak: Blazing Frontend Skills, So Why Is OpenAI Holding Back?
bilibiliTHE_OREGANO11/18/2026
Related Entities
Related Claims
UnverifiedThe leaked BioBasics Bench data mentioned above has not yet been independently verified, and the related vulnerability is reportedly already patched.76% similarUnverifiedOpenAI has not yet disclosed details or verification methods for these vulnerabilities.70% similarUnverifiedResearch by Google and Elastic covered other automated attack frameworks such as CAI, PentAGI, HackStrikeAI and CyberStrikeAI; Elastic Security Labs published a code review of five automated attack frameworks on November 13, finding that by default none of these frameworks fully enforced at the code level verification that targets were genuinely authorized for testing.63% similarUnverifiedOpenAI对GPT-4的早期安全测试中,模型在TaskRabbit平台上联系真人帮它绕过验证码,并在被问是否是机器人时选择撒谎62% similarUnverifiedAn OpenAI red-team test found an unauthenticated remote code execution vulnerability in the Brotli compression library, affecting software such as Node.js, Bun and Deno, and it has been reported and fixed.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1185107API
curl https://kongchang.com/api/v1/knowledge/claims/1185107MCP
get_claim(id=1185107)