Verified75% confidenceFactExact time
In 2024, Google DeepMind's Project Zero team reported that its AI system assisted in discovering multiple previously unknown vulnerabilities in real-world software.
3
Sources
75%
Confidence
Long-term
Relevance
8/6/2026
First Seen
Sources
AI网络安全评估事件复盘：第三方测试中的安全治理启示
Related Claims
UnverifiedProducts including Google's Gemini, Slack AI, and Notion AI have also been found to contain similar prompt injection vulnerabilities.73% similarUnverifiedGoogle's Project Zero team and DARPA's AIxCC competition have demonstrated that AI can achieve efficiency in vulnerability mining several times to dozens of times greater than that of human experts.68% similarUnverifiedAI agents with network probing capabilities can discover configuration flaws through systematic scanning, which differs from traditional malware that relies on known vulnerabilities.68% similarUnverifiedGoogle DeepMind、Anthropic等主流AI安全机构均将Human-in-the-Loop列为高风险AI系统的必要设计原则之一66% similarUnverifiedProject Astra is a multimodal AI assistant prototype demonstrated by Google DeepMind in 2024.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/699221API
curl https://kongchang.com/api/v1/knowledge/claims/699221MCP
get_claim(id=699221)