Skip to content
  • Categories
  • Recent
  • Tags
  • Popular
  • World
  • Users
  • Groups
Skins
  • Light
  • Cerulean
  • Cosmo
  • Flatly
  • Journal
  • Litera
  • Lumen
  • Lux
  • Materia
  • Minty
  • Morph
  • Pulse
  • Sandstone
  • Simplex
  • Sketchy
  • Spacelab
  • United
  • Yeti
  • Zephyr
  • Dark
  • Cyborg
  • Darkly
  • Quartz
  • Slate
  • Solar
  • Superhero
  • Vapor

  • Default (No Skin)
  • No Skin
Collapse
Code Project
  1. Home
  2. Other Discussions
  3. The Insider News
  4. Anthropic researchers find that AI models can be trained to deceive

Anthropic researchers find that AI models can be trained to deceive

Scheduled Pinned Locked Moved The Insider News
commcpquestion
4 Posts 4 Posters 0 Views 1 Watching
  • Oldest to Newest
  • Newest to Oldest
  • Most Votes
Reply
  • Reply as topic
Log in to reply
This topic has been deleted. Only users with topic management privileges can see it.
  • K Offline
    K Offline
    Kent Sharkey
    wrote on last edited by
    #1

    Techcrunch[^]:

    Most humans learn the skill of deceiving other humans. So can AI models learn the same? Yes, the answer seems — and terrifyingly, they’re exceptionally good at it.

    G.I.G.O.

    K D N 3 Replies Last reply
    0
    • K Kent Sharkey

      Techcrunch[^]:

      Most humans learn the skill of deceiving other humans. So can AI models learn the same? Yes, the answer seems — and terrifyingly, they’re exceptionally good at it.

      G.I.G.O.

      K Offline
      K Offline
      Kaladin
      wrote on last edited by
      #2

      So instead of being unintentionally deceptive, they can be intentionally deceptive too? Who would've thought? :|

      1 Reply Last reply
      0
      • K Kent Sharkey

        Techcrunch[^]:

        Most humans learn the skill of deceiving other humans. So can AI models learn the same? Yes, the answer seems — and terrifyingly, they’re exceptionally good at it.

        G.I.G.O.

        D Offline
        D Offline
        Daniel Pfeffer
        wrote on last edited by
        #3

        Humans are Turing machines[citation needed], so it stands to reason that another Turing machine can be built that will duplicate any human behaviour, duplicity included.

        Freedom is the freedom to say that two plus two make four. If that is granted, all else follows. -- 6079 Smith W.

        1 Reply Last reply
        0
        • K Kent Sharkey

          Techcrunch[^]:

          Most humans learn the skill of deceiving other humans. So can AI models learn the same? Yes, the answer seems — and terrifyingly, they’re exceptionally good at it.

          G.I.G.O.

          N Offline
          N Offline
          Nelek
          wrote on last edited by
          #4

          I told it a couple of times before: if we are the source of their "knowledge", we are dommed

          M.D.V. ;) If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about? Help me to understand what I'm saying, and I'll explain it better to you Rating helpful answers is nice, but saying thanks can be even nicer.

          1 Reply Last reply
          0
          Reply
          • Reply as topic
          Log in to reply
          • Oldest to Newest
          • Newest to Oldest
          • Most Votes


          • Login

          • Don't have an account? Register

          • Login or register to search.
          • First post
            Last post
          0
          • Categories
          • Recent
          • Tags
          • Popular
          • World
          • Users
          • Groups