Anthropic's Mythos Model Is Being Accessed by Unauthorized Users

6 comments

• lccerina 7 days ago
  On other news, the AI company recently found to have poor cybersecurity posture, still has poor cybersecurity.
• htrp 7 days ago
  How is this even possible? You'd think you would use a frontier-tier cybersecurity model to secure itself.
  [-]
  • stared 7 days ago
    Mythos can secure all models that cannot secure themselves. Can Mythos secure itself?
    [-]
    • skeledrew 6 days ago
      Hmm if Mythos can secure all models that can't secure themselves, then it's likely that Mythos can secure itself. So since it can secure itself, then it can't secure itself because it can.......
• ineedasername 7 days ago
  >"A handful of users in a private online forum gained access to Mythos on the same day that Anthropic first announced a plan to release the model to a limited number of companies for testing purposes, said the person, who asked not to be named for fear of reprisal. The group has been using Mythos regularly since then, though not for cybersecurity purposes, said the person, who corroborated the account with screenshots and a live demonstration of the model..."
• eur0pa 6 days ago
  It's oddly quiet in here
• xvxvx 7 days ago
  So… the AI model deemed ‘too dangerous to release’, that the US government is using… had unauthorized users?

  I’m starting to think the AI bubble wont so much burst as explode.