9 pointsby vincent_s5 hours ago2 comments
  • robertknight4 hours ago
    This post is full of Claude's irritating verbal ticks ("It's not X, it's Y"). It is fine to use Claude (or another LLM) to help with this kind of reverse engineering, but I would prefer the write-up be done by hand.
  • d-lo4 hours ago
    There's another good write up and discussion on Reddit (with fewer LLM writing ticks): https://old.reddit.com/r/ClaudeCode/comments/1ujilqt/anthrop...

    I do wonder if Anthropic is modifying output via prompt modification or some of the Fable style weights adjustments for requests that contain these sentinel values. That would be one way to try to prevent distillation, and they have shown a willingness to silently modify model behavior for user input they deem dangerous.

    • samuelknight2 hours ago
      The Reddit poster called it spyware which is not right. They are using stenography to fingerprint accounts using Claude Code instances routing through Chinese resellers.
    • Chu4eeno3 hours ago
      They have always sent a ton of analytics + everything you do, is this news just because the Chinese are afraid they'll get cut off?