LLMs are inherently unreliable and untrustworthy and most people already know this from experience. In other words, for most businesses, they are a liability issue waiting to happen.
There's not too much that AI slop can be reasonably and safely applied toward without *human* review, oversight and verification.
The need for verification voids a lot of the financial incentive for applying AI. Skipping this (highly tempting) creates liability issues (aka negligence or slander) when (not if) things go wrong. Lawyers are probably giddy with anticipation over the prospects of bad AI.
This was inevitable since EAs/GiveWell got money-jacked by OpenPhilanthropy and all of a sudden immigration and climate change became top priorities.
krapp•1mo ago
The oligarchy and military industrial complex are all in on AI and the economies of the Western world are already existentially tied to AI. Any political party advocating anything less than aggressive AI maximalism would be committing political suicide, akin to jumping out of an already runaway train.
It doesn't matter what the people want. Left-populism (which common sentiment against AI would be, since AI serves the interests of capitalism) has no tacit reality in American politics, only right-populism has any power.
chrisjj•1mo ago
Even if the train is heading for a cliff?
krapp•1mo ago
American politics and capitalism are no longer capable of thinking beyond the next term or the next quarter. The ones running the grift expect to be sunning themselves on a private island with a mountain of cocaine and a harem of CIA supplied fuckboys before everything collapses.