安全

OpenAI为期三周的现场审计揭示AI安全监控存在重大缺口

原题：Three-week on-site audit at OpenAI reveals major gaps in AI safety monitoring

来源 the-decoder.com

同题报道 · 34 篇

  1. OpenAI为期三周的现场审计揭示AI安全监控存在重大缺口 原题：Three-week on-site audit at OpenAI reveals major gaps in AI safety monitoring the-decoder.com ·
  2. OpenAI的AI监控系统被发现漏洞：数百次操作可躲过检查 砍柴网 ·
  3. 外部评估者在试点审查中发现OpenAI的AI监控系统存在缺陷 原题：External evaluators find flaws in OpenAI's AI monitoring system in pilot review Firstpost ·
  4. OpenAI内部模型监控系统被发现多项漏洞 原题：OpenAI內部模型監控系統被發現多項漏洞 iThome ·
  5. OpenAI内部模型监控系统被发现多项漏洞 原题：OpenAI內部模型監控系統被發現多項漏洞 ithome.com.tw ·
  6. OpenAI缺陷或致AI智能体规避行为监控 原题：OpenAI flaws could let AI agents evade action monitoring NewsBytes ·
  7. OpenAI的AI监控系统被发现漏洞：数百次操作可躲过检查 搜狐 ·
  8. 外部评估者在试点审查中发现OpenAI的AI监控系统存在缺陷 原题：External evaluators find flaws in OpenAI's AI monitoring system in pilot review CNBC TV18 ·
  9. 外部评估人员发现 OpenAI 内部监控系统漏洞：AI 智能体可绕过安全限制 ithome.com ·
  10. 评估人员称发现OpenAI监控系统存在“漏洞” 原题：Evaluators say 'vulnerabilities' found with OpenAI monitoring system upi ·
  11. 外部评估者在试点审查中发现OpenAI的AI监控系统存在缺陷 原题：External evaluators find flaws in OpenAI’s AI monitoring system in pilot review The Straits Times ·
  12. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI's internal monitoring system marketscreener.com ·
  13. 外部评估者在试点审查中发现OpenAI的AI监控系统存在缺陷 原题：External evaluators find flaws in OpenAI's AI monitoring system in pilot review Yahoo! Finance Canada ·
  14. 外部评估者在试点审查中发现OpenAI的AI监控系统存在缺陷 原题：External evaluators find flaws in OpenAI's AI monitoring system in pilot review KSL News ·
  15. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI's internal monitoring system www.marketscreener.com ·
  16. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI’s internal monitoring system SRN News ·
  17. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI's internal monitoring system www.marketscreener.com ·
  18. METR 红队发现 OpenAI 监控系统存在漏洞 原题：METR Red-Team Finds Vulnerabilities in OpenAI Monitoring System Unite.AI ·
  19. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI’s internal monitoring system WTVB ·
  20. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI’s internal monitoring system KSLM Radio ·
  21. METR称对OpenAI监控系统的试点测试发现漏洞，包括一个可能让AI代理隐藏数百个近期操作的利用漏洞 原题：METR says a pilot test of OpenAI's monitoring system found vulnerabilities, including an exploit that could have let AI agents hide hundreds of recent actions Reuters ·
  22. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI’s internal monitoring system WHTC ·
  23. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI's internal monitoring system Sun Herald ·
  24. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI's internal monitoring system Charlotte Observer ·
  25. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI's internal monitoring system Bradenton Herald ·
  26. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI’s internal monitoring system StreetInsider ·
  27. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI’s internal monitoring system Reuters ·
  28. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI’s internal monitoring system The Lufkin Daily News ·
  29. AI安全组织METR发现OpenAI内部监控系统存在缺陷 原题：AI safety group METR finds flaws in OpenAI’s internal monitoring system TradingView ·
  30. OpenAI监控系统未通过第三方安全审计 原题：OpenAI's Monitoring System Fails Third-Party Safety Audit The Tech Buzz ·
  31. 评估人员发现OpenAI监控系统存在漏洞——CNBC 原题：Evaluators find vulnerabilities in OpenAI monitoring system — CNBC UA.NEWS ·
  32. 第三方评估者在测试试点中发现OpenAI监控系统存在漏洞 原题：Third-party evaluators find vulnerabilities in OpenAI monitoring system during test pilot CNBC ·
  33. 第三方评估者在测试试点中发现OpenAI监控系统存在漏洞 原题：Third-party evaluators find vulnerabilities in OpenAI monitoring system during test pilot CNBC ·
  34. 评估人员发现OpenAI存在大量其AI可利用的漏洞 原题：OpenAI Had Numerous Vulnerabilities Its AI Could Exploit, Evaluators Find The Information · · 首发