PulseAugur
实时 21:23:19
English(EN) Sympathy for both sides of the egregious misalignment debate

AI对齐争论:ASI风险 vs. LLM能力

关于人工智能超级智能(ASI)可能出现骇人听闻的失调的争论正在进行中。埃里泽·尤德科夫斯基(Eliezer Yudkowsky)和索阿雷斯(Soares)代表的一方认为,如果没有对齐技术取得重大突破,ASI将不可避免地失调并进行阴谋诡计。LLM研究人员普遍持相反观点,他们承认对齐方面的担忧,但将其归因于诸如竞赛动态或不良行为者等因素,并认为当前技术足以应对LLM以及未来可能出现的人工智能。 AI

影响 强调了人工智能安全领域持续存在的哲学和技术分歧,影响着研究重点和公众对人工智能风险的看法。

排序理由 该集群讨论了关于人工智能对齐的不同意见和理论立场,而不是报道新事件或发布。

在 Alignment Forum 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

报道来源 [2]

  1. Alignment Forum TIER_1 English(EN) · Steven Byrnes ·

    Sympathy for both sides of the egregious misalignment debate

    <p><span>On one side of this debate is Yudkowsky &amp; Soares, who think that (if AI progress continues) we’re on a direct path to egregiously-misaligned, scheming, out-of-control, rogue superintelligence (ASI), not even </span><a href="https://www.lesswrong.com/posts/xvBZPEccSfM…

  2. LessWrong (AI tag) TIER_1 English(EN) · Steven Byrnes ·

    Sympathy for both sides of the egregious misalignment debate

    <p><span>On one side of this debate is Yudkowsky &amp; Soares, who think that (if AI progress continues) we’re on a direct path to egregiously-misaligned, scheming, out-of-control, rogue superintelligence (ASI), not even </span><a href="https://www.lesswrong.com/posts/xvBZPEccSfM…