I have no objections against Meta employees fapping on company time.

So what? Isn’t that some inner-Meta work related shit? Lowest-manager tier problem?

So?

This article is more fitting for the science community than here.

That’s the funniest part here: those LLS are not very effective, but still “AI” companies have an enormous amount of money.

Technology!

HoloGuard: don’t touch holograms, you holohobo!

Shares-shmares. More engineering, less shit-trading!

0.076

Update when you can afford downtime. The update should be painless but you know how it is: things go crazy when they shouldn’t. Every time.

Yes, only curated, selected and censored garbage from now on! :)

The same way as using non-English makes it harder to find any data manually in books/forums/etc(except for strictly local data like laws or similar).

Almost everyone does.

– You share when you share, yes.
– But I said that only to (pathetically enormous list of recipients).
– What is known to 2 persons is known to pig.

In the Sinclair era we used cassettes instead of discs. It was shitty but cheap solution.

Unification and standardization. 5 models. Intricate tuning is done by customer PC style.

Forget it, you people are so accustomed to being corpo-slaves that you don’t deserve a dictator like me :)

Models, not factories. I don’t care what factory would produce them. Plus the concept of “competition and price” is an abstract thing from economics textbooks. You know, like “communism” or those kids buying a crazy numbers of watermelons. In the real world “competitors” would form cartel and regulate prices as they wish. And please, don’t counterargument with a reference to something that happened in '53. That is no more relevant than late Roman economics from II century CE :)

If I were a dictator I would permit only 4-5 models of smartphones. Why produce huge amounts of shit if you can just make a few decent machines? That would save an enormous amount of resources and energy.

But without widespread usage there won’t be any breakthroughs… Anti-LLM folk could redirect their attention to the core problem (energy source, wind and solar shouldn’t raise any concerns) but they prefer to attack the LLMs themselves. Early implementations of every technology were crazy energy ineffective at the moment when becoming practically useful.

In some areas the interview process became so crazy that passing the interview is a rather separate knowledge. Using LLMs is an adequate answer to this ridiculousness.

People could use “common sense” class. Many have problems with deciding who/what is trustworthy enough. Plus some basic explanation of how LLMs work in principle might help tremendously with that overtrust issue.