7 pointsby 1vuio0pswjnm74 hours ago2 comments
  • jqpabc1234 hours ago
    These sort of "investigations" are just getting started.

    AI is marketed as being "smart" but marketing hype can't hide the fact that it is also prone to being random and irrational and illogical and simply untrustworthy.

    Vendors are faced with the difficult task of selling a blatantly flawed tool for big $.

    AI in it's current form is a liability issue waiting to happen and industry insiders know it. Yann Lecun has said that he thinks LLMs are nearing their end.

    https://www.newsweek.com/nw-ai/ai-impact-interview-yann-lecu...

    • cyanydeeza few seconds ago
      also, theres no solution to it because it emmulates a human well enough that it can commit many criminal offenses without normal human liabilities.

      its uber-next: break laws and hope society just accepts.