this post was submitted on 20 Jan 2024
50 points (83.8% liked)

Free Open-Source Artificial Intelligence

2908 readers
1 users here now

Welcome to Free Open-Source Artificial Intelligence!

We are a community dedicated to forwarding the availability and access to:

Free Open Source Artificial Intelligence (F.O.S.A.I.)

More AI Communities

LLM Leaderboards

Developer Resources

GitHub Projects

FOSAI Time Capsule

founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
[–] Buffalox 15 points 10 months ago* (last edited 10 months ago) (3 children)

Is Zuckerberg an idiot? Or does he have an actual plan with this?
Seems to me it's completely useless like Metaverse.
If the LLM is so stupid it can't figure out the sides of an equal sign can be reversed as simple as in 2+2=4 <=> 4=2+2. He will never achieve general intelligence by just throwing more compute power at it.
As powerful as LLM is, it's still astoundingly stupid when it hits its limitations.

[–] [email protected] 12 points 10 months ago (2 children)

Humans are astoundigly stupid when they hit their limitations.

[–] art101 4 points 10 months ago

Spoken like a true AI!!!! ;-)

[–] Sylver 0 points 10 months ago (2 children)

The difference is that we can go beyond that limitation. Even self-coding AI will either solve a problem, or compound its own inefficiencies before asking an operator to help out.

[–] [email protected] 7 points 10 months ago (1 children)

Your post sounds almost as dense as:

"everything that can be invented has been invented." - Duell 1899.

[–] Sylver 3 points 10 months ago

The difference here is that Zuck is not planning on inventing or revolutionizing anything. He’s just throwing more computation power at an already inefficient method of modeling AI.

[–] [email protected] 3 points 10 months ago

or compound its own inefficiencies before asking an operator to help out.

Some people do. Some people refuse to ask for help.

[–] deafboy 6 points 10 months ago (2 children)

I don't know much, but from what I know, we still haven't reach a point of diminishing returns, so more power = more better.

[–] fidodo 2 points 10 months ago

Not necessarily since you also need better techniques. A competitor could easily surpass you with less by being smarter about how the AI is trained.

[–] fidodo 5 points 10 months ago* (last edited 10 months ago) (1 children)

Trying to achieve AGI by throwing more compute at LLMs is like trying to reach the moon by building a more powerful hot air balloon.

[–] Buffalox 2 points 10 months ago (1 children)

Assuming that "not compute" should be "more compute" I totally agree. That's a very apt analogy.

[–] fidodo 2 points 10 months ago* (last edited 10 months ago)

Yes, thanks, swipe typing picked up "not" instead of "more". Maybe someone can throw some more compute at the swipe typing algorithm to better pick up on the context of the sentence when picking words.