How rational is Eliezer Yudkowsky’s prophecy?You’ve probably seen this one before: first it looks like a rabbit. You’re totally sure: yes, that’s a rabbit! But then — wait, no — it’s a duck. Definitely, absolutely a duck. A few seconds later, it’s flipped again, and all you can see is rabbit.The feeling of looking at that classic optical illusion is the same feeling I’ve been getting recently as I read two competing stories about the future of AI.According to one story, AI is normal technology. It’ll be a big deal, sure — like electricity or the internet was a big deal. But just as society adapted to those innovations, we’ll be able to adapt to advanced AI. As long as we research how to make AI safe and put the right regulations around it, nothing truly catastrophic will happen. We will not, for instance, go extinct.Then there’s the doomy view best encapsulated by the title of a new book: If Anyone Builds It, Everyone Dies. The authors, Eliezer Yudkowsky and Nate Soares, mean that very literally: a superintelligence — an AI that’s smarter than any human, and smarter than humanity collectively — would kill us all.
pull down to refresh
related posts
132 sats \ 6 replies \ @freetx 19h
They might have a point if their pattern matching algo was sentient or could even learn by itself. But it cant.
The "AI" industry regularly engages in these "investment farming" tactics (ie. oh wow look at how dangerous our new tech is, imagine how much it must be worth).
We really don't have the right terminology yet for what they've built. Its not technically "AI", but its certainly useful. Its something like a dynamic search engine, but thats not nearly as exciting to invest in as the "this tech is nation-state level critical that may wipe out the species!" sales pitch.
reply
18 sats \ 5 replies \ @optimism 16h
Do you feel that the LLM as we know it today is (or could be in a followup iterarion) as big a game changer as electricity or the public internet as quoted above?
I still have deep doubts. Sure, it can hurt search engines immensely, but I won't really cry over google losing part of its revenue. The disruptor eventually becomes a sweet incumbent target for disruption. Cyber-darwinism.
reply
111 sats \ 4 replies \ @freetx 16h
To be honest I debate that as well. I think it will have less of an impact than internet but more of an impact than say "Google" or "Social Media".
The "hell" I worry about is not being gunned down by sentient AI drones, but rather losing any ability to find a decision maker in any given context.
Imagine the hell we feel now trying to call a MegaCorp (ie. Cable company) to get some problem resolved....now multiply that by 100.
Your debit card will stop working one day and you wont be able to talk to anyone to find out why or what you can do to fix it.
reply
21 sats \ 2 replies \ @optimism 15h
Currently I rate it "as bad an impact as social media" but only for the ready-made consumer apps, because what these really do is establish addiction for the masses, just like FB/Twitter. The business integrations are in some cases better but many people I've chatted with lately are on the negative sentiment; that it's more trouble than it's helping ultimately.
Very good point. Plus any managers won't even know the policies they are supposed to manage as those were generated, so they'll just say "no". It'll be a great time being a lawyer I guess! Laughing all the way to the bank gets a 100x too.
reply
211 sats \ 1 reply \ @freetx 15h
The one benefit is maybe when we call MegaCorp instead of the soulless automated voice system we have now, we can be told our questions are "Amazing" and our complaints have "Excellent Insight!"
Then finally... when the AI bot ultimately can't solve the problem maybe we can witness it go into a schitzo doom loop and finally admit to us that it utterly failed and should be permanently turned off.
reply
21 sats \ 0 replies \ @optimism 14h
Lol!
If only we had the 3 laws of robotics baked into these bots...
reply
18 sats \ 0 replies \ @fourrules 15h
Hell is administered by dumb bots
reply
100 sats \ 0 replies \ @SimpleStacker 19h
I responded to Yudkowsky's ideas here. I don't think there's a high chance of human doom due to AI development.
#1009710
reply