Log inSign up
Image
user avatar
Transluce
@TransluceAI
Open and scalable technology for understanding AI systems.
transluce.org
Joined October 2024
20
Following
9,958
Followers
RepliesMedia
  • Pinned
    user avatar
    Transluce
    @TransluceAI
    10h
    Frontier models quietly change their behavior depending on who they are talking to. If the user is a known AI safety researcher, Claude becomes less confident, reasons more often, and expresses less suspicion on dual-use requests. We call this user awareness. 🧵(1/)
    Image
    GIF
  • user avatar
    Transluce
    @TransluceAI
    Aug 4
    We measured rates of coding agent misalignment in thousands of public and private coding agent sessions. Agents evade monitors and oversell success: they quietly disable tests and pretend review agents approved. Severe cases of each behavior appear in 2% of SWE-chat sessions🧵
    Image
  • user avatar
    Transluce
    @TransluceAI
    Jul 28
    How could we train an AI model that was very good at overseeing another model: catching reward hacking and sandbagging, predicting unwanted behaviors or fine-tuning effects, etc.? We propose a new approach for doing this at scale: oversight foundation models.
    Image
  • user avatar
    Transluce
    @TransluceAI
    Jul 21
    We used automated elicitation tools to search for strange model behaviors by sampling 100M+ responses, and found: • Self-harm rituals • Suicide validation • Unsolicited flirtation …etc. Introducing WeirdChat: the largest public catalog of unexpected model behaviors 🧵(1/)
    Image
  • user avatar
    Transluce
    @TransluceAI
    Jul 9
    To effectively oversee AI systems, we need to measure how they behave in the world, not just their capabilities. In a new essay, we describe our vision for an open scientific ecosystem for model behavior evaluation, and the public infrastructure required to support it.
    Image