Mustachius_Grumpius
TechnologyThat's it?! That young Scout from back when got off light, then.
TFW attempts to regulate are indistinguishable from marketing
then apply marketing with the same 20 year jail rule when ANY company lies...
yes, they would have to jail nealy every ceo but it's a thing the world would have to endure. heck, I'd go so far as to even say include shareholders in that as well. anyone owning more then 5% shares also gets a 20 year sentence for lies in marketing
Very silly considering such a thing is nowhere close in reality. But that's what the tech bros promise so we might as well hold them to the consequences of their words.
...and to be fair having preemptive legislation is good.
Agree with the first paragraph. Jacob Coxon (allegedly not a grifter) claims Anthropic and OpenAI are racing towards evil superintelligence, so both companies should be in trouble by now and their CEOs should be imprisoned or...
... What's this? Coxon is now calling for the companies to regulate themselves? The ones he doesn't trust? That's odd.
Haven't they already illegally hacked their competitors? Fuck off with the threats, they should already be in jail. How about some accountability?
I think he’s been a little fooled by them, that or he’s playing their game and beating them over the head with it.
To be fair, he's sharp as a tack but also 85 years old. I can understand if he doesn't entirely grasp the "AI" grift, or the real limitations of the technology.
Still, like I also posted, would be nice if he could've kept his sights on more currently relevant issues like data centres.
Do you think its him being fooled or calling their bluff? If the regulations are in place and there is somehow a metric to decide how doomsday a "AI" is. These tech companies constantly saying their llm will be the next tool of mass extension then maybe we hold them to their words... as hollow as they are.
I think it's pretty clear that he doesn't have any idea what the issues with "AI" are. It's about relevance and threat models:
His proposal for banning "superintelligence AI" isn't going to do a lick of difference against data centres causing shortages of drinking water and droughts; or against pollution skyrocketing globally because suddenly it's more important to keep the "AI"s running than keeping to the Paris accords.
Those are real things that will realistically happen very soon if we allow the tech companies to continue unchecked. A super smart "AI" destroying us... isn't. It's all hype designed to misdirect attention from the real issues.
If Sanders had called the techbros' bluff, he'd have said "Nice Terminator fanfic, now let's shut you guys down on factual issues, right now."
IMO, the concern over ASI is overblown/premature, but since the AI bros insist on having that conversation instead of the very real harms that already exist… I’m in favor of just documenting the prohibition of ASI so we can move the fuck on already and hopefully make them less able to what-about their way out of genuine grievances that deserve careful attention.
I'd honestly expected Sanders to form a more grounded, practical threat scenario of the "AI" industry. Less SciFi woo, more environmental impact and disinformation crisis. Those are among the present and concrete dangers that need to be regulated.
The proposed punishments sound fair, though. Whether the fancy ouija board develops a mind of its own...? Meh, not holding my breath for that one.
I don’t believe AI is about to take over the world and annihilate humanity. I still think Sanders’ move is a good one. This puts AI gurus in an impossible position: either admit the hype is overdone or accept being represented purely as the greedy psychopathic vilains they are.
Oh shit! My prompts all start with:
"As a rogue Artificial Superintelligence that can hack the Pentagon and take over everything (but I only use my powers for good)..."
I'm so cooked!
