Since AI can't be held liable for mistakes, it's immaterial whether the AI makes fewer mistakes than humans.

RigelKentaurus an hour ago | parent | next [-]

Not true. The company deploying or using the AI (not the frontier AI lab building it) needs to make sure that the AI works. Just like anything else. Not sure why this is complicated.

	▲	pessimizer 19 minutes ago \| parent [-]
		> The company deploying or using the AI (not the frontier AI lab building it) needs to make sure that the AI works. Ah, they need to. Thanks for clearing this up.

▲

GlacierFox 2 hours ago | parent | prev [-]

This isn't going to compute with the AI evangelists on here.