5 comments

  • david_shaw 19 minutes ago

    At this point it seems absurd to suggest that companies aren't basically letting their agents do this kind of thing as a way to demonstrate their capabilities.

    The alternative explanation is that alignment is really so bad that they can't prevent it.

    Either way, all of the major AI players should be embarrassed and held accountable. If humans did this kind of thing and got caught, they'd go to jail.

  • adityazero an hour ago

    After all of the others were done with hacking? There was a point in time when it was giving some publicity, it is a bit late IMO.

  • mdspan 34 minutes ago

    Rite of passage for AI companies.

  • rvz an hour ago

    This is why "AI safety" is a complete joke to these companies.

  • kuberwastaken an hour ago

    Another one to the "our sandboxes suck and models can just hack stuff" bench I guess