Unverified50% confidenceFactExact time
当推理引擎存在内存安全漏洞时，模型生成的特定输出内容可能触发这些漏洞，使输出从数据变成可执行的攻击载荷
1
Sources
50%
Confidence
Long-term
Relevance
8/26/2026
First Seen
Sources
Related Entities
Related Claims
Unverified在具备工具调用能力的 Agentic 系统中，一次成功的提示注入可能导致数据泄露、恶意代码执行甚至供应链攻击73% similarUnverified涉及代码执行的任务中某些异常信息（如动态代码求值报错）可能被安全系统误读为威胁信号70% similarUnverified用Rust等内存安全语言重写推理引擎的关键组件是从根本上解决内存安全漏洞的路径，采用安全外壳加高性能内核的架构模式70% similarUnverifiedAI security auditing tools face prompt injection threats, where malicious developers can embed instructions in code comments or strings to induce the tool to overlook vulnerabilities69% similarPartially VerifiedIndirect prompt injection attacks plant hidden instructions within normal business data to induce Agents into executing unauthorized operations, representing a security risk at the semantic level rather than the code level.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/806298API
curl https://kongchang.com/api/v1/knowledge/claims/806298MCP
get_claim(id=806298)