Guardian Angels: LLM Personalization for Productivity and Security

 (gwern.net)

41 points | by andsoitis 9 hours ago

5 comments

  • nshotton 45 minutes ago
    I might be missing something here, but could this concept not also be used in the same way for harm? For example if a model is trained to replicated a sucessfull scammer?
  • great_wubwub 53 minutes ago
    IDK, this seems like the next step towards uploading yourself to the cloud.
  • Avicebron 1 hour ago
    Why digital twins? What about the NetNavis from Megaman, just aligned to the users interests?
  • Uinside 31 minutes ago
    [dead]
  • ai_fry_ur_brain 1 hour ago
    [dead]