隐藏指令操纵司法AI，危及公正性
英文摘要
A ThinkBig Blog report warns that prompt injection attacks can manipulate AI systems used in judicial processes through hidden instructions. Such manipulation could secretly alter legal decisions, compromising the right to a fair trial and legal security. Experts and judges are increasingly concerned about this invisible threat to judicial impartiality.
中文摘要
ThinkBig博客的一篇报道警告称，提示注入攻击可通过隐藏指令操纵司法过程中使用的AI系统。这种操纵可能秘密改变法律裁决，损害获得公正审判的权利与法律安全。专家和法官对这种无形威胁对司法公正性的侵蚀日益担忧。
关键要点
Prompt injection enables attackers to embed hidden instructions that manipulate AI outputs in judicial contexts.
提示注入使攻击者能够嵌入隐藏指令，操纵司法语境中的AI输出。
Such manipulation could secretly alter legal proceedings, undermining fair trials and legal security.
这种操纵可能秘密改变法律程序，破坏公正审判和法律安全。
Legal experts and judges are increasingly alarmed by this emerging threat to impartiality.
法律专家和法官对这一威胁公正性的新兴风险日益警惕。