Its a supply and demand problem. The supply of human AI gig workers vastly staggeringly exceeds the demand for human AI workers. Trash talking the entire indu

Its a supply and demand problem. The supply of human AI gig workers vastly staggeringly exceeds the demand for human AI workers. Trash talking the entire industry just means the poor will get poorer and be worse treated when the demand dries up even more. So I'm not really clear on the social justice angle of the article. This does bring up an important long term market point. If there are an essentially infinite number of people willing to drive a car for practically nothing, tell me again why do we need self driving cars as an example of AI? Its not like we're running out of poor people anytime soon, they don't cost much of anything, so why waste all that money on kicking the poor people out of the economy? Eventually, once everyone's kicked out of the economy, there won't be anyone left to buy the very expensive self driving cars anyway, and no one will need self driving cars because they'll be nowhere economically viable left to drive to. Certainly we can "self reproduce" car drivers much cheaper than we can produce self driving cars full of unobtainable microprocessors, and producing human drivers is much more "black swan" proof than producing globalist microprocessors. The other problem is a economy, in the long run, has to serve its participants. Lets say as a thought experiment its inevitable that all capital and all economic activity will solely be concentrated in those with IQs over 150 or sociopathic tendencies or outright criminality, and of course with everyone else kicked out of the economy they will be the sole beneficiaries of AI, much like the European Royalty was, in some sense, the the only beneficiary of gold mining industry. Well... won't 99% of the population go all French Revolution on them and fire up the guillotines? As soon as those people at the top of the pyramid are removed, and there won't be too many of them, the 99% of the population can go back to happy productivity. Any individual who rebels during the runup will get steamrolled by the rest of their competitors who don't rebel from the plan, but the endpoint of the plan is anyone cooperating with the plan to the inevitable endpoint will get the guillotine in the end, so what's the maximally efficient game theory perspective if you're trapped in a self destructive game where either total participation or obvious non-participation results in your demise? Well, everyone's better off if everyone "phones it in" and fails. So I expect a lot of "totally unexpected" AI failures. Another philosophical problem is we've run a repeated massively parallel experiment on producing smarter better educated "non-artificial" intelligence entities known as the university system, etc. It really hasn't worked out other than the usual primate dominance ritual purposes and making certain the people at the top stay at the top. But people seem to think if we emulate that nonsense that never worked in a computer, it'll work next time. Sort of like perpetual motion doesn't work IRL out in the real world but if you abuse a CAD/CAM program hard enough a simulation of perpetual motion will work in a computer ... but even if we fake it in a computer simulation, that STILL doesn't mean it'll work IRL. Its almost like a "heisenberg uncertainty principle" of AI where if we had AI it seems the universe routes around it such that it can't make a net profit. Kind of like burning two barrels of crude oil to grow one barrel of ethanol to replace a barrel of crude oil, graph that out and see where it leads? I'm just saying, if "low IQ" biologicals are already economically excluded, why would "low IQ" AIs not also be economically excluded? And we've proven we can't educate "better" biologicals, so trying to produce a master race of AIs is likely to fail just as badly as those past experiments.