Dell-grundlægger siger rolig om SHODAN-stil Sentient AI, fordi 'du husker ozonlaget og det hele', og det fik vi rettet

Genudgivet af Platon

Abonnenter: 0

Vi lever i en æra med AI-hype, og alle har et bud. Men selvom de fleste af os er en smule bekymrede over, hvad fremkomsten af ultraprædiktiv tekst betyder for menneskelig kreativitet og kritik, bekymrer nogle få Silicon Valley-typer sig om Artificial General Intelligence eller AGI, som dybest set er et seriøst klingende udtryk for selvlærende AI med sans og potentielt et uudsletteligt begær efter menneskeblod. Eller noget i den stil.

Men Dells grundlægger og administrerende direktør Michael Dell siger, at du ikke skal bekymre dig. I en nylig virtuel brandchat med formueforvaltningsfirmaet Bernstein (spottet af Registret), Dell said that he worried about the advent of AGI “a little bit, but not too much.” Why? Because “For as long as there’s been technology, humans have worried about bad things that could happen with it and we’ve told ourselves stories… about horrible things that could happen.”

That worrying, continues Dell, lets humanity “create counter actions” to prevent those apocalyptic scenarios from playing out before they happen. “You remember the ozone layer and all,” said Dell to Bernstein’s Tony Sacconaghi, “there are all sorts of things that were going to happen. They didn’t happen because humans took countermeasures.”

Dell (the man) went on to say that Dell’s (the company) AI business was booming. “Customer demand nearly doubled quarter-on-quarter for us and the AI optimized backlog roughly doubled to about $1.6 billion at the end of our third quarter,” beamed Dell (the man again), which—and I write this as someone for whom ‘literally GLaDOS’ ranks low on the list of fear priorities—does seem like the kind of thing a tech CEO would say in the prologue to a film about AI killing everyone.

Regardless, Dell reckons you shouldn’t be worried about the robot uprising any time soon, because humans are just that good at recognising and heading off problems before they occur. Except for that climate change thing and the nanoplast i vores blod, I guess. Oh, and the fact that we didn’t start fixing the ozone layer until there was already a gaping hole in it (that won’t be fixed until 2040, or 2066 if you happen to live in the Antarctic). If you’ll permit me a bit of editorialising, which I guess I’ve already been doing, that feels like reaching the right conclusion for the wrong reasons.

For my money, you shouldn’t worry about AGI because it’s a spooky story well-off tech types dreamt up to hype up the capabilities of their faktiske AI tech and because it’s a much neater and easier tale to cope with than the things which are really scary about AI: the potential for the decimation of entire creative industries and their replacement by homogenous robotic sludge. Plus, the possibility that the internet—for all its problems, a genuinely useful repository of human knowledge—becomes a great library of auto-completed and utterly incorrect nonsense of no use to anyone.

After all, I’ve already reached the point where I append most of my Google searches with “Reddit” to make sure I’m actually getting human input on whatever problem I’m facing. And that’s a much trickier problem with much more profit-threatening solutions than is the bogeyman of HAL 9000.