News for hackers. Stuff that matters.
Re:Adversarial Policies Beat Superhuman Go AIs by bubaumba on Tuesday December 24, 2024 @10:35AM (#42502548)
> I don't think I see them as a win

Unavoidable, probably

> but they're easily dealt with. AI will need analysts at the latter stage to evaluate the outputs but that will be a relatively short-lived problem

That solves only to some degree. Hallucinations may happen at this stage too. Then either correct answer can get rejected or false pass through.

There are no comments on this comment yet.