Unverified60% confidenceEventExact time
In 2023, the White House organized an AI red-teaming challenge in which thousands of participants stress-tested models from multiple companies.
2
Sources
60%
Confidence
Long-term
Relevance
8/9/2026
First Seen
Sources
AI模型协同利用漏洞：大模型训练中的安全隐患解析
Related Claims
Unverified2023年白宫组织了大规模AI红队活动，邀请数千名研究人员对多个前沿模型进行安全评估77% similarUnverifiedThe AI executive order issued by the White House in 2023 requires safety testing (Red Teaming) of foundation models.76% similarUnverifiedAt the 2023 DEFCON security conference, multiple AI laboratories jointly held a large-scale public red team test.65% similarUnverifiedOS World基准测试由学术界于2024年提出，专门测试AI模型在真实操作系统环境中完成复杂任务的能力63% similarUnverifiedGAIA基准测试由Meta AI和HuggingFace等团队于2023年提出，包含400多个精心设计的问题61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/716935API
curl https://kongchang.com/api/v1/knowledge/claims/716935MCP
get_claim(id=716935)