Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Meta's Vision for Superintelligence

meta.com

Post reply on HN
Print view View original

41–50 of 184 posts

Re: Meta's Vision for Superintelligence

#41

Post by FirmwareBurner »

sorcerer-mar wrote: post #21
colesantiago wrote: post #2

I don't buy these Superintelligence claims from Meta of 'abundance' and 'humans are free to do other things' as they are tied to Wall St and earnings. It is always abundance for the super rich, scarcity for those in jobs. How can I be free to do my gardening whenever I want when the landlord is asking for $11K rent in my SF flat? So eventually they will do the opposite of this 'vision' and put this super intelligence…

> How can I be free to do my gardening whenever I want when the landlord is asking for $11K rent in my SF flat? This is the fatal flaw. It's been recognized explicitly for at least 140 years that the price of land rent rises in lockstep with productivity increases, guaranteeing there is no "escape velocity" for the labor class regardless of how good technology gets.

>regardless of how good technology gets

Technology increases aren't there so you work less hours for the same pay, they're there so your business owner gets more money form you working the same hours.

If a machine gets invented that can do your job it's not like you can now go home and relax for the rest of your life and still keep receiving your pay cheques. This utopia doesn't exist.

Re: Meta's Vision for Superintelligence

#43

Post by jameskilton »

Facebook / Meta and Mark in particular are an amazing case of someone who is, at least at this point, incapable of learning from past mistakes, or even recognizing the mistakes they have made and are continuing to make.

Facebook's mission of "connecting the world" turned out to be the absolute worst thing anyone should ever try to do. Humans are social creatures, yes, but every connection we make costs energy to maintain, and at a certain point (Dunbar's Number) we apply the minimal amount of energy and effort. With Internet anonymity, that means we are actually incapable of treating each other as people on the Internet, leading to the rise of toxicity and much, much worse.

Mark has never understood this, and as his fortune is built around not understanding this, he never will.

There is nothing good that will come from Meta's "superintelligence" and this vision is proof.

Re: Meta's Vision for Superintelligence

#44

Post by aschobel »

Barrin92 wrote: post #6

Anthropic, to their credit, just published that pretty funny paper about how they couldn't get their chatbot to run a single vending machine profitably[1]. Admirable that Mark is so visionary he's skipped straight past Bodega-intelligence to super-intelligence [1] https://www.anthropic.com/research/project-vend-1?ref=blog.m...

I wonder if that is the right takeaway; that was Sonnet 3.7

Model since then have been able to run it profitably. Incredible how fast things are progressing.

https://andonlabs.com/evals/vending-bench

Re: Meta's Vision for Superintelligence

#45

Post by lvl155 »

hnthrow90348765 wrote: post #34

This stuff is naive. There's a bunch of people who want a large income (wealth) disparity, and they will fight to preserve it unless you give them an equivalent station in the 'new world'. But you will still need to sustain ex-workers if they can't get normal jobs, and those same people at the top will not tolerate the taxes required to sustain a basic level of living for much wider population. They already can't tol…

Last time I commented on Zuck on HN I got a warning. That said, this stuff confirms for the 100th time that he is out of touch with reality. Perhaps that’s why he wanted to make VR work so badly. I think he might try to change Co’s name again since Meta doesn’t fit the bill anymore. How about they buy Intel and reverse merge just for the name?

Re: Meta's Vision for Superintelligence

#47

Post by shortrounddev2 »

The effect of Meta products from 2006~2022 has been disastrous for humanity. Social media companies are the new big tobacco. Allowing them to play a leading role in the development of AGI would probably be fatal for democracy around the world. I would not trust a single word that comes out of Mark Zuckerberg's mouth

Re: Meta's Vision for Superintelligence

#48

Post by coldpie »

techpineapple wrote: post #28
coldpie wrote: post #11

Deeply embarrassing to see this coming from one of America's most powerful companies, especially after the metaverse faceplant from a few years back.

Why embarrassing? Having one failure means you can’t try again?

LLMs are not going to fundamentally reshape society. Productivity will go up a couple percent, the profits from that productivity will go to the richest people. Some jobs will be made redundant, some new jobs will be created. That's it. Same as it ever was.

Re: Meta's Vision for Superintelligence

#49

Post by abxyz »

“As recently as 200 years ago, 90% of people were farmers growing food to survive. Advances in technology have steadily freed much of humanity to focus less on subsistence and more on the pursuits we choose”

More people today are dying from starvation than people existed on earth 200 years ago. Celebrating our achievements in making shareholders rich is one thing, but to take credit for freeing the people. Yikes. Mark is more out of touch than seems possible.

Re: Meta's Vision for Superintelligence

#50

Post by qprofyeh »

Reads as RIP Metaverse to me.

Any time a CEO publishes such empty, wordy essays, it's probably earnings reporting time. I can't shake the feeling it's a public subreply at one of or a cluster of doubting investors, who started to doubt the CEOs vision for the company, or find the lack of one on a certain topic concerning.

Post reply on HN