1 Matching Annotations
  1. Last 7 days
  2. www.technologyreview.com www.technologyreview.com
    https://www.technologyreview.com/2026/06/05/1138437/the-meta-hack-shows-theres-more-to-ai-security-than-mythos/
    1
    1. fxp007 05 Jun 2026
      As AI models continue to improve, hardening their defenses might actually get easier.

      大多数人认为随着AI能力增强，安全挑战会越来越大，但作者认为更先进的AI模型实际上可能使防御更容易。这个反直觉观点挑战了人们对AI安全发展的线性认知，暗示AI进步可能同时带来更强大的防御能力，而非仅仅增加攻击面。

      counterintuitive ai-evolution security-defense
    Visit annotations in context

    Tags

    Annotators

    URL

    technologyreview.com/2026/06/05/1138437/the-meta-hack-shows-theres-more-to-ai-security-than-mythos/