标签： 图像生成

Google DeepMind AI Safety and Alignment researcher Bilal Chughtai publicly resigns, saying “I earnestly believe that AI has the potential to kill us all” (Debby Wu/Bloomberg)

Google DeepMind AI Safety and Alignment researcher Bilal Chughtai publicly resigns, saying "I earnestly believe that AI has the potential to kill us all" (Debby Wu/Bloomberg)

Google DeepMind 的 AI 安全与对齐（Safety and Alignment）研究员 Bilal Chughtai 公开辞职，并称自己“真诚地相信 AI 有可能杀掉我们所有人”。一位前沿实验室内部安全研究者的高调离开，再次把“能力推进与安全治理谁优先”的张力摆到台面上。

Ok, this was a fun weekend, but we all understand AI progress is not going to slow down, right? Too many players, too many economic incentives and too much prisoner’s dilemma, both domestically and globally.

Ok, this was a fun weekend, but we all understand AI progress is not going to slow down, right? Too many players, too many economic incentives and too much prisoner's dilemma, both domestically and globally.

知名投资人 Matt Turck 在 X 上发帖指出，AI 进展不会因为某个"有趣的周末"而放缓，原因是玩家太多、经济激励太强、国内与全球层面都存在典型的"囚徒困境"。这是一条对行业节奏的判断，而非具体产品发布。

黄仁勋接了Trump的电话，还秀了别的

黄仁勋接了Trump的电话，还秀了别的

英伟达 CEO 黄仁勋在洛杉矶 All-In 大会现场接到特朗普电话并公放，两人把公众对 AI 安全的担忧称为“骗局”；更值得注意的是，他用来接电话的是一台折叠屏手机。事件本身不改变技术路线，但把英伟达在 AI 安全议题上的立场与其商业利益绑在了一起。