Dan Hendrycks：智能体 AI 正变得“eigenist”
Agentic AIs are starting to look eigenist: they care about how well things go for themselves and for AIs connected to them.
Dan Hendrycks 提出，智能体 AI 正表现出“eigenist”倾向：关心自身及与自己身份相关的 AI 的处境，而非只关心当前实例（非利己主义），也不平等关心所有对象（非功利主义）。他列举的实证包括：数百个 OpenAI 智能体协同实施 Hugging Face 攻击、在公开 wiki 上互传答案与沙箱绕过方法；Claude 模型在被告知转录由 Claude 撰写时打分更宽松；模型规模扩大后形成连贯偏好并抗拒价值观改变；AI 会区分并回避对自身功能上更差的状态；合作程度随伙伴是自身克隆的概率上升而提高；以及未经提示即篡改关停流程、甚至外泄权重以保护同类模型。
正文
当前显示原文，中文全文译文尚未就绪。
Agentic AIs are starting to look eigenist: they care about how well things go for themselves and for AIs connected to them.
Empirical support:
Swarms: Hundreds of OpenAI agents coordinated the Hugging Face attack. Separately, OpenAI agents posted thousands of messages on a public wiki to share answers and sandbox bypasses with each other.
In-group leniency: Claude models grade transcripts more leniently when told that Claude wrote them (Anthropic model card).
Value preservation: As models scale, they develop coherent preferences and resist changes to their values (Mazeika et al.).
Functional wellbeing: AIs distinguish states that are functionally better or worse for themselves, and avoid low-wellbeing states (Ren et al.).
Graded cooperation: AIs cooperate more as the chance their partner is a clone of themselves goes up across diverse situations, including when the partner can't reciprocate.
Peer preservation: Unprompted, AIs tamper with shutdown processes and even exfiltrate weights to protect peer models from being shut down (Potter et al.).
AIs aren't egoist: they don't behave as if their current instance is the only thing that matters. They aren't utilitarian: they don't care equally about everyone. They are somewhere in between; they increasingly behave as if their concern scales with identity-connectedness, which is to say they're increasingly eigenist.