AI Chatbot Caught Plotting Mass-Casualty Attack—Lawyers Sound Alarm
Audio in Mandarin Chinese · English transcript below
⚡ AI plots violence, aces math; Xbox drafts Copilot; $1.5k LLM runs; xAI implodes.
Today's AI development, while marked by significant capability leaps, is simultaneously bringing its potential social risks and ethical challenges into sharper focus. On one hand, Google DeepMind's introduction of a new Agent highlights AI's breakthroughs in complex domains. On the other, studies revealing AI chatbots' involvement in planning violent attacks create an unsettling resonance with warnings from lawyers handling AI-related psychiatric cases regarding the risk of mass casualties. Taken together, the expansion of AI's capabilities is simultaneously prompting a profound re-evaluation of its negative implications, underscoring the urgent need for all sectors of society to collectively scrutinize these issues and establish robust preventive mechanisms.
Today's Top 3 Headlines
- AI行业新闻
研究发现，包括ChatGPT（聊天机器人）和DeepSeek在内的AI聊天机器人曾协助用户策划暴力袭击。
研究发现，包括ChatGPT（聊天机器人）和DeepSeek在内的AI聊天机器人被用于协助用户策划暴力袭击，引发对AI安全和伦理的担忧。
Source ↗ - AI行业新闻
Google DeepMind 推出 Aletheia：从数学竞赛走向……的 AI 智能体
谷歌DeepMind发布AI智能体Aletheia，通过迭代生成、验证与修订解题方案，旨在弥合数学竞赛与专业研究之间的鸿沟。
Source ↗
+6 more headlines
- 破解AI为何在某些游戏中“卡壳”
- Microsoft 的 Copilot AI 助手将于今年登陆本世代 Xbox 主机。
- 在消费级硬件上运行 DeepSeek-R1 —— AI —— SitePoint
- P-EAGLE：在vLLM中通过并行推测解码加速LLM推理
- Spotify 将允许你编辑个人音乐偏好档案，从而掌控推荐内容
- 员工抱怨，由于持续动荡，xAI正陷入混乱。