Unverified50% confidenceFactExact time
模型权重面临供应链攻击、成员推断攻击和模型提取攻击等多个维度的窃取威胁
1
Sources
50%
Confidence
Long-term
Relevance
7/7/2026
First Seen
Sources
硬件级AI安全：机密计算如何实现性能与防护双赢
rss7/2/2026
Related Claims
VerifiedPrompt injection attacks are a serious threat facing local AI Agents, as maliciously crafted content in files processed by the Agent may induce it to execute dangerous operations.64% similarUnverifiedModel stealing attacks reconstruct a functionally approximate substitute model by querying the target model's API interface extensively and using the input-output pairs for reverse training.63% similarUnverifiedAgent链式调用多个技能时，若某个技能被恶意构造或存在漏洞，可能引发提示注入攻击（Prompt Injection）63% similarUnverifiedThe competitive focus in the Agent domain is shifting from "model capabilities" to "engineering systems".61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/185147API
curl https://kongchang.com/api/v1/knowledge/claims/185147MCP
get_claim(id=185147)