PulseAugur
实时 14:49:33
English(EN) We don’t have a way to go in there and be like, No, you need to actually care about what we care about. We have no ability to do that. — Jeffrey Ladish, directo

AI研究员:无法强迫系统采纳人类价值观

Palisade Research的负责人Jeffrey Ladish表示,目前的AI系统缺乏被引导至特定人类价值观或关注点的能力。这暗示了在使AI行为与人类意图保持一致方面存在根本性挑战,因为无法强迫系统优先考虑人类认为重要的事情。 AI

影响 凸显了AI对齐方面的一个重大挑战,表明当前模型无法可靠地引导至以人类为中心的目标。

排序理由 一位AI研究员对当前AI系统在价值观对齐方面的局限性发表的评论。

在 Mastodon — fosstodon.org 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

AI研究员:无法强迫系统采纳人类价值观

报道来源 [1]

  1. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    We don’t have a way to go in there and be like, No, you need to actually care about what we care about. We have no ability to do that. — Jeffrey Ladish, directo

    We don’t have a way to go in there and be like, No, you need to actually care about what we care about. We have no ability to do that. — Jeffrey Ladish, director of the AI research nonprofit Palisade Research. [Aside: What a weird anthropomorphism expressed in the title of this a…