Filip HNov 11
nixCraft 🐧
Meta's chief AI scientist, Yann LeCun, plans to exit and launch his own start-up. He ended up in a position where he has to report to a kid much younger than him. Yann LeCun, who is a scientist, strongly maintains the stance that AI will never be able to reason and plan like humans. He said LLM has its uses, but it is next to impossible to build AGI or even intelligence that can match a baby elephant's intelligence. No wonder Zuck is not happy 🤣 https://www.ft.com/content/c586eb77-a16e-4363-ab0b-e877898b70de
Client Challenge

Nov 11 at 5:57pmWeb
Show threadnixCraft 🐧Nov 11
The truth is, all billionaires want to play God with their AGI projects. Most of these billionaire brats do not have a scientific background. They think throwing GPUs and a large data center with younger engineers is going to build the next big thing: AGI. In that AGI world age, everyone is jobless and they are controlling everything. Unfortunately, their chief scientist does not believe anything like that is remotely possible. That's why they forced him to exit.
Show threadRichard NairnNov 11
@nixCraft Once you understand that LLMs are statistical then you understand why AGI is impossible. It isn’t creative. It can’t think outside the box - the box defines it. General intelligence requires advanced knowledge and rule interaction that AI can’t reproduce.
Show threadSnuffleupadonNov 11
@RichardNairn @nixCraft the things to me is that when anyone says "AI" today they really mean LLMs whereas previously they meant Machine Learning. It's a semantic thing whereby the powers that be want us conflating what's really going on.
Show threadTor Iver WilhelmsenNov 11
@Unsightly3055 @RichardNairn @nixCraft Machine learning, neural nets, inference engines, Prolog and Lisp, the AI world was so much more fun before it turned into stochastic lying.
Show threadFloNov 11

@RichardNairn
They are not just statistical, but also interpolating models. They are therefore particularly bad at extrapolating. Different approaches might yield more „creative“ results in the future, but the current approaches are uniquely bad at it. (At least that’s my understanding)

@nixCraft

Show threadJan AntošNov 11
@nixCraft I can see a lot of use cases for AI (Summarization, etc…) but not in any way AGI, even replacement of programmers I see impossible. Just ask any of those Code Helping AI agents to implement QR generator from scratch without using any existing lib.
Show threadlemgandiNov 11

@nixCraft This reminds me so much of the Government Cryptography 'debate'.

JUST NERD HARDER DAMMIT

Show threadMadalin Ignisca 🇪🇺🇪🇦Nov 11

@nixCraft so one billionaire manages to fund and have it's AGI built and running. What is next? Manages to create everything for everybody? Everybody becomes jobless? He realizez that nobody can buy what he creates? He becomes Saint AGI and gives "the machine" for nothing in return to humanity?

Well, we all know how it ends when "The Machine Stops".

Show threadKévin ⏚Nov 12

@madalinignisca @nixCraft that’s kind of the point though. There is this weird ideology they have to collapse society (acclerationism), hide in their little bunkers for a while, then slither out to become Gods and shape the place as they want.

Of course to anybody sane this doesn’t make sense, mainly as they’ll probably end up being killed by their security teams either before going underground or within the first few weeks, as they’ll do anything but be nice to them

Show threadL'égrégore André ꕭꕬNov 12
@madalinignisca @nixCraft I don't think the tech bro funded ones ever will go past "The Machine Slops"
Show threadWaitman GobbleNov 12
TESCREAL - Wikipedia
TESCREAL is a neologism proposed by computer scientist Timnit Gebru and philosopher Émile P. Torres. An acronym, it stands for Transhumanism, Extropianism, Singularitarianism, (modern) Cosmism, Rationalists (the internet community, not to be confused with other uses of the term), Effective Altruism, and Longtermism.[1][2] Gebru and Torres argue th...
TESCREAL - Wikipedia

Show threaddodothedev🦤💻Nov 11

@nixCraft
Wow, this guy sounds bright, what was he doing at Meta?

Its nice to hear that he has his own views and isn't just parroting what the TechBro's say to stay in a job. I bet working at Meta must have been like talking to a brick wall.

I hope to see him succeed.

Show threadFer24 🇲🇽Nov 11
@nixCraft
So the retarded of Zuckerberg strikes again
Show thread@iveylineNov 12
@nixCraft Like the Metaverse Zucks is facing znother reality check with AI.
Show threadCD CastilloNov 16
@nixCraft isn’t he one of those that got promised a 100 million per year??

I guess that salary came in shares (Monopoly money) and with conditions.