> Have you ever seen someone claiming ai labs should have no liability in the wild? I haven't

I'd describe David Sacks's position as, approximately, no limits on AI. At some hypothetical future state, sure, maybe, but right now, nothing. That's tantamount to consequence-free action.