4 ms·

ChatGPT can give plausible problematic scenarios where you (admittedly) cannot. Reflect on that.

• sfn42 3y ago
  Not really sure what you're saying. The LLM is capable of parroting doomsayings from text it's trained on? Yeah that does not come as any kind of surprise. Maybe you want to bring up a point, something concrete that we can actually talk about? Because this vague appeal to authority (where the "authority" is a notorious liar) isn't very convincing.
  • z5h 3y ago
    I’m suggesting you’re not a position to make claims about a system, that in its infancy, can outperform you (most of us) in many ways. Especially if, in your opening argument, you admit being unable to think of scenarios instead of doing the work to conceive of plausible scenarios.
    • sfn42 3y ago
      [flagged]