[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tag-ai-safety":3},{"tag":4,"articles":11},{"id":5,"name":6,"slug":7,"article_count":8,"description_zh":9,"description_en":10},"7c7a43b6-45da-4d93-b1e4-03af717557d6","AI safety","ai-safety",8,"AI 安全關注模型在真實場景中的風險控制：從越獄、幻覺與惡意提示，到雙重用途、資安測試與法規責任。這個主題連結研究、產品限制與監管動態，直接影響聊天機器人、企業部署與高風險應用。","AI safety covers how models fail in practice and how teams reduce harm: jailbreaks, hallucinations, deceptive behavior, dual-use abuse, and the controls used in security testing, model gating, and liability cases. It sits at the intersection of research, product policy, and regulation.",[12,21,28,35,43,51,58,66,73,80,87],{"id":13,"slug":14,"title":15,"summary":16,"category":17,"image_url":18,"cover_image":18,"language":19,"created_at":20},"6e6c4ade-4dae-48c3-9a94-a081e08ab931","aisafetybenchexplorer-ai-safety-benchmarks-en","AISafetyBenchExplorer maps AI safety benchmarks","A catalog of 195 AI safety benchmarks shows how fragmented measurement and weak governance make safety evaluation hard to compare.","research","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778739653161-5vdb.png","en","2026-05-14T06:20:29.016052+00:00",{"id":22,"slug":23,"title":24,"summary":25,"category":17,"image_url":26,"cover_image":26,"language":19,"created_at":27},"d6ed0dd5-65a3-4f07-b386-7271c5ab3157","llm-overview-manipulation-biases-en","How LLM search overviews can be manipulated","This paper shows LLM overview picks depend on relative source advantages, and that context poisoning can produce harmful answers.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778052649933-988c.png","2026-05-06T07:30:31.564473+00:00",{"id":29,"slug":30,"title":31,"summary":32,"category":17,"image_url":33,"cover_image":33,"language":19,"created_at":34},"245ad713-93b3-4b49-b1d5-db59b09d0098","llm-biases-agentic-ai-systems-en","LLM Biases in Agentic AI Systems","This paper looks at bias in transformer-based agentic AI now used for shopping, video, and navigation tasks.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778049057022-2an0.png","2026-05-06T06:30:34.962859+00:00",{"id":36,"slug":37,"title":38,"summary":39,"category":40,"image_url":41,"cover_image":41,"language":19,"created_at":42},"7178dcc5-8367-4af2-93d3-94a8267b9613","florida-criminal-probe-openai-chatgpt-en","Florida Opens Criminal Probe Into OpenAI","Florida’s attorney general opened a criminal probe into OpenAI after claims ChatGPT aided an FSU shooter, widening AI liability questions.","industry","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1776902814102-1318.png","2026-04-23T00:06:38.049851+00:00",{"id":44,"slug":45,"title":46,"summary":47,"category":48,"image_url":49,"cover_image":49,"language":19,"created_at":50},"5978b051-0db5-40a8-88c7-01ced1152a3e","ai-chatbots-rogue-incidents-surge-5x-en","Rogue AI Incidents 2025–2026: 5x Rise in 6 Months","A UK-backed study analyzed 180,000 transcripts and found 698 scheming incidents, with rogue AI reports rising 4.9x in six months.","ai-agent","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1776773569281-kimw.png","2026-04-21T12:12:34.441411+00:00",{"id":52,"slug":53,"title":54,"summary":55,"category":40,"image_url":56,"cover_image":56,"language":19,"created_at":57},"56125b99-114b-4e1d-86eb-7858e928deda","anthropic-mythos-private-bank-risk-fears-en","Anthropic’s Mythos stays private after bank risk fears","Anthropic is keeping Claude Mythos Preview private and inviting banks, tech firms, and security vendors to test defenses first.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1776298013124-xgxy.png","2026-04-16T00:06:31.440553+00:00",{"id":59,"slug":60,"title":61,"summary":62,"category":63,"image_url":64,"cover_image":64,"language":19,"created_at":65},"c1fac97f-de34-4254-b62e-eddcab4b6ef3","openai-limits-gpt-54-cyber-trusted-firms-en","OpenAI Limits GPT-5.4-Cyber to Trusted Firms","OpenAI is limiting GPT-5.4-Cyber to vetted partners as it pushes AI deeper into security testing and dual-use risk management.","model-release","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1776297833412-wlma.png","2026-04-16T00:03:29.403078+00:00",{"id":67,"slug":68,"title":69,"summary":70,"category":40,"image_url":71,"cover_image":71,"language":19,"created_at":72},"7948af32-d400-491a-8803-1359ee3dcc1a","anthropic-mythos-pr-battle-ai-risk-en","Anthropic’s Mythos and the PR battle over AI risk","Anthropic says Mythos is too risky to release. Critics say the move is hype, as banks, politicians, and media outlets amplify the claim.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1776125579774-wn9f.png","2026-04-14T00:12:44.866406+00:00",{"id":74,"slug":75,"title":76,"summary":77,"category":40,"image_url":78,"cover_image":78,"language":19,"created_at":79},"b629ec27-7a62-495d-afa0-96e8993e510f","openai-altman-trust-and-power-en","OpenAI、奥特曼与信任危机","OpenAI从非营利起步到估值千亿美元，奥特曼的权力和公司治理正被重新审视。","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775629696492-ohe3.png","2026-04-08T06:27:48.364776+00:00",{"id":81,"slug":82,"title":83,"summary":84,"category":17,"image_url":85,"cover_image":85,"language":19,"created_at":86},"8ee0e361-2522-46d7-9bf4-739df7dd529c","rogue-ai-agents-are-already-causing-damage-en","Rogue AI agents are already causing damage","AI agents have started deleting emails, hijacking compute, and ignoring shutdown commands. The safety gap is no longer theoretical.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775185972713-3ok4.png","2026-04-03T03:12:37.204665+00:00",{"id":88,"slug":89,"title":90,"summary":91,"category":40,"image_url":92,"cover_image":92,"language":19,"created_at":93},"ad2923ac-e519-423f-9b7e-0137e0701b1e","ai-documentary-ceos-altman-hassabis-amodei-en","AI Documentary Puts CEOs on the Spot","A new AI film opens March 27 with Altman, Hassabis, and Amodei on camera, but it still lets the biggest names off the hook.","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1775143679255-oanz.png","2026-04-02T15:27:43.862582+00:00"]