[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-ai安全":3},{"tag":4,"articles":11},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":10},"6b9f373e-17e2-4fbd-bf9c-e7b81e2412ae","AI安全","ai安全",5,"AI安全涵蓋模型失控、內容濾網、提示注入、資料外洩與實體威脅等面向，重點在於如何評估風險、降低誤判，並在產品上線前建立可驗證的防護與治理流程。","AI safety covers model misbehavior, content filtering, prompt injection, data leaks, and even physical threats around the industry. For developers, it matters because shipping AI requires measurable risk controls, review pipelines, and governance that can hold up in production.",[12],{"id":13,"slug":14,"title":15,"summary":16,"category":17,"image_url":18,"cover_image":18,"language":19,"created_at":20},"07968cc8-f216-4938-9e22-1b7f92ae3311","claude-code-leak-reveals-hidden-features-en","Claude Code泄露后，代码库露出什么","Claude Code源码意外外泄后，社区挖出51万行代码、26个隐藏指令和6级安全验证，细节很扎眼。","tools","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775179135628-e89m.png","en","2026-04-03T01:18:34.353501+00:00"]