3 ms·
"I feel fear about an impending doom. Yudkowsky's argument that a superintelligent AI will inevitably destroy humanity seems to have no flaw. Yet, nobody seriou
by kittikitti 20d ago
"I feel fear about an impending doom. Yudkowsky's argument that a superintelligent AI will inevitably destroy humanity seems to have no flaw. Yet, nobody seriously tries to sandbox AIs because they are too useful with access."
No one takes Yudowsky's claims that AI will destroy humanity seriously. There are many people who are seriously trying to sandbox AIs.
- jackb4040 20d ago"Nobody" if you're democratically-minded and think in terms of number of people. But the handful of tech billionaires at the helm of the world economy absolutely do believe it. Or at least pretend to believe it. Or maybe they can't even tell the difference anymore as they strip-mine science fiction for any vision of a future that doesn't just look like an incrementally worse version of the present.
- ben_w 20d ago> No one takes Yudowsky's claims that AI will destroy humanity seriously. There are many people who are seriously trying to sandbox AIs. I know some people at AI labs, there's definitely people working in this field who take it seriously. This does not stop them also working on e.g. better sandboxes.