Verified75% confidenceFactExact time
Jailbreak attacks refer to bypassing a model's safety alignment mechanisms through carefully crafted prompts to make it output harmful content.
3
Sources
75%
Confidence
Long-term
Relevance
7/5/2026
First Seen
Sources
Related Claims
UnverifiedPrompt injection attacks can be divided into two categories: direct prompt injection and indirect prompt injection, with indirect prompt injection being more harmful because the attack surface encompasses all external data sources reachable by the agent.78% similarUnverified模型逃逸攻击通过添加大量感叹号或无意义特殊字符欺骗模型绕过安全限制77% similarUnverified通用越狱指单一提示词或操作序列能系统性绕过模型所有安全约束，而针对性越狱只能解锁某一类具体有害行为77% similarUnverified过度授权（如在只需生成文档时同时开放命令执行权限）会因提示词注入攻击或模型幻觉引入破坏性操作的安全风险77% similarUnverified模型层安全约束（对齐训练、系统提示词、输出过滤器）都面临越狱风险，精心构造的提示词可以绕过这些约束72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/112159API
curl https://kongchang.com/api/v1/knowledge/claims/112159MCP
get_claim(id=112159)