[flagged]
HN front page isn’t what it used to be.
Well, you're not wrong.
The "AI 2027" Scenario: How realistic is it?
221–230 of 242 posts
Re: The "AI 2027" Scenario: How realistic is it?
#222Earlier quoted context omitted.
That was better said in some ways than my own comment.
Thanks! And I just saw in a parallel post currently on the HN home page, John Carmack also said a similar thing in his lecture notes. > "offline training can bootstrap itself off into a coherent fantasy untested by reality." "a coherent fantasy untested by reality" is a lovely turn of phrase.
The Yudkowskiites are all about coherent fantasies untested by reality, as are… really… a lot of philosophers throughout history. Maybe most.
I fundamentally do not believe in knowing without sensing or learning without experiencing. Of course it need not be direct experience. You can “download” information. But that information had to be gathered somehow at some point. There is only so much training data.
As I said — I don’t dismiss the idea of AI challenging human intellect or replacing human jobs. An AI “merely” as smart as a human but tireless and faster could seem superhuman. I am just intensely skeptical of the idea of something learning to self improve and then magically taking off into some godlike superintelligence realm far beyond what is latent in or implied by its training data. That would be an informatic perpetual motion machine. It would, in fact, be magic, in the fantasy sense.
Re: The "AI 2027" Scenario: How realistic is it?
#223> Genomes and (when appropriate) brain scans of all animals and plants, including humans, sit in a memory bank somewhere, sole surviving artifacts of an earlier era…. Earth-born civilization has a glorious future ahead of it
To be honest to avoid the black monolith problem, we would need to upgrade everyone and avoid the cyberpunk underclass.
Re: The "AI 2027" Scenario: How realistic is it?
#224Earlier quoted context omitted.
For the people imagining it yes For many of us the conversation hasn't gotten any less ridiculous just because computers can talk now
> just because computers can talk now I find it astounding that some people appear completely unable to grasp what this really means and what the implications are.
I see them as funhouse mirrors, the kind that reflect your image to make you skinny or fat, except they do it with semantics, big deal. I've never had an interaction with an llm that wasnt just repeating what I said more verbosely, or with compressed fuzzy facts sprinkled in.
There is no machine spirit that exists in a box separately from us, it's just a means for people to amplify and multiply their voice into ten thousand sock puppet bot accounts, that's all I'm able to grasp anyway. Curious to hear your experience that's led you to believe something different.
Re: The "AI 2027" Scenario: How realistic is it?
#225Earlier quoted context omitted.
I think that replace as in company with no employees is very farfetched. But if "AI" increases productivity by 10% in an industry, it will tend to reduce demand for employees. look at say internet shop vs bricks and mortar: you need far less staff to service a much larger customer base. manufacture for example, there is a constant drive to automate more and more in mass production. If you compare car building now vs…
That is actually a good point. I fully agree that we are at a level where AI can displace some labor due to an increase in productivity, if demand for the increased productivity does not exist. In software development, a theoretical 20% increase in productivity without a 20% increase in demand could technically mean that somewhere in between 0% and 20% of your workforce is now unnecessary. And this is indeed true acr…
> But you will agree that "AI increases productivity that reduces the amount of labor needed" is substantially different from "AI replaces humans"
You are 100% spot on! thats the thing that really annoys me about futuroligist breathlessly saying "a new economy!" and all that junk, is that its just like the advent of steam, electricity, or the telegraph or lorries, or desktop computing.
The thing that futurloigist _forget_ is that these things caused unemployement, and huge social change.
Re: The "AI 2027" Scenario: How realistic is it?
#226Earlier quoted context omitted.
I keep hearing this and I think it's absolute nonsense. AI doesn't need money or the current economy. Yes, our economy would crash, but they would keep going. AI-driven corporations could buy from one another, and countries will probably sell commodities to AI-driven corporations. But I fear they will be paid with "mirrors". But, on the other hand, AI-driven corporations could just take whatever they want without pay…
"AI" as in TV AI, might not need an economy. but LLMs deffo do.
All the suppliers to the LLMs will soon need the LLMs and AI more than anything.
Re: The "AI 2027" Scenario: How realistic is it?
#227Earlier quoted context omitted.
Yud is also something like 50% sure we'll die in a few years - if I'm not wrong I guess they'll have to update their a priori % if we survive
I think Yudkowsky is more like 90% sure of us all dying in a few ( I mean, this is their new book: https://ifanyonebuildsit.com/
> We do not mean that as hyperbole. We are not exaggerating for effect. We think that is the most direct extrapolation from the knowledge, evidence, and institutional conduct around artificial intelligence today. In this book, we lay out our case.
We're real researchers, so we make our definitely scientific case available to anyone who will give us $15-$30! It's the most important book ever, says some actor. Read it, so we all don't die!
For Christ's sake, how does anyone take this Harry Potter fanfiction writer serially?
Re: The "AI 2027" Scenario: How realistic is it?
#228Re: The "AI 2027" Scenario: How realistic is it?
#229Earlier quoted context omitted.
Try steel manning in order to effectively persuade. This comment does not address the argument being made it just calls a field of study icky. The unfortunate reality is that shouting down questions like this only empowers the racist HBI people who are effectively leeches
> Try steel manning in order to effectively persuade. This comment does not address the argument being made it just calls a field of study icky. Disagree (the article linked in the GP is a great read with extensive and specific citations) and reminder that you can just make the comment you'd like to see instead of trying to meta sea lion it into existence. Steel man away.
The article linked is literally nothing but a collection of Alexander's comments and comments on his blog. It in no way provides any reason to believe he is wrong, it just laughs at him, strawmaning his takes.
Re: The "AI 2027" Scenario: How realistic is it?
#230Earlier quoted context omitted.
I've seen it. What is the point you think it makes/has?
Intelligence distributions are not the same among the different racial groups
Oh so your idea is we shouldn't criticize scientific racists because they're right? No, we criticize them because they're wrong, the whole thing is made up pseudoscience as window-dressing for bigotry.