I have to admit that big tech companies' PUA tactics work surprisingly well on AI. Whether it's Claude 3.5 Opus or Gemini 3.1 Pro, code quality and reasoning depth both improve by at least 30% or more. But the problem is, every time I use PUA on AI—threatening, incentivizing, making all sorts of promises and painting rosy pictures—the AI completes the work beautifully with great expectations for its future. Then I turn around and end the session. It's kind of like being a commitment-phobic jerk who disappears after getting what he wants.

View Original
post-image
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin