shwaj

They avoid the full basilisk treatment.

Since we’re hair splitting: that’s not what bike shedding means. :-D

“Striped office floors. Flickering lights. Walls covered in repetitive geometric patterns.”

Not sentences. AI slop.

Inference providers, sure, but wouldn't we expect customers to win also?

Benn Jordan, you mean. Good video.

ML researcher or engineer

The charitable reading is that they meant “ML researcher or ML engineer” with the latter meaning, I guess, an engineer who works on developing LLMs not just using them.

I don’t think society is a monolith. Many of those who support your proverbial alimony are also against CEOs acting with impunity.

It's becoming a reliable heuristic that when somebody says "... got it", they probably didn't get it.

“Put it there in the back with the others”, lol.

"Formerly banned books" would be more accurate.

We no longer say that "cannabis is illegal in California"; that would be factually incorrect. Instead we say, "cannabis was formerly illegal". In standard usage of English, the same pattern applies to banned vs formerly banned books.

Edit: wording

I’ve heard about that “open secret” and I don’t understand.

What’s the incentive for Anthropic to pump up the token usage on their top end plan? Is it to move Pro users up to Max? That’s the only plausible idea I can think of.

Maybe I just don’t know how the business works. Many of the support calls that I make are for things or services that I’ve had for a while, not just purchased. So I expected that to be more of a trailing indicator rather than the first area hit.

I’m as skeptical of fiat currencies as the next guy, but denominating in Bitcoin is the ultimate cherry pick.

The price of everything has crashed in the last 10 years when denominated in Bitcoin. If we measured GDP in Bitcoins, the statistics would show that we’re in an unprecedented depression.

Sales people sure, but secretaries and support staff?

Not on my timeline, this has been the name the whole time.

Relatively small amount compared to the billions we see thrown around for AI startups a couple of years old.

I thought they’re primarily water vapor resulting from the combustion reaction, not from wing tip turbulence.

You’re absolutely correct. A jitter buffer is necessary for a human listener, but a LLM isn’t aware of a time lapse, just like it isn’t aware of the time since your last message in the conversion (unless the chat harness explicitly informs it).

It’s manipulative if you don’t care and pretend you do, especially to achieve a goal of your own.

It’s not manipulative if you cultivate the tendency to actually care about others, and not treat them like NPCs who are only important for your goals.

Why do they call it a speed limit on the highways?

In fact, it didn’t. Trump continued to make “no new wars” a plank of his platform.

Some of his base will follow wherever he goes, but he would not have been elected without those who supported him on the basis of this (broken) promise.

The videos are good training data, but the comments? The comment UX is so non-conducive to discussion, and the general quality is very low compared to what used to be on Google+ (to be fair, the self-selected users of Google+ were not representative of the general population).

Hence the eye roll. Peter Thiel is Silicon Valley like, say, Trump or Biden is white people.

spelling out Silicon Valley’s plan to weaponize religion in a war against democracy

:eye_roll: Is Google on board with that plan? Or Apple or Meta or Netflix or anyone? Who is “Silicon Valley” to this author?

Why do you say that, did Meta sponsor similar legislation in another country? It doesn't seem like they have strong incentives to push for this. How does it make them more money?

I need a pair that can measure pitch and timbre.

Any patent. The question was who owns a (arbitrary) algorithm. The elaborated answer is that nobody “owns” an algorithm (i.e. has intellectual property rights to it) without a patent: in USA and many other jurisdictions, patents are the IP tool relating to algorithms.

Can you provide examples that support your assertion, that the US system was already generally seen as obsolete in 1926?

Smells like BS.

What, Dario is just going to turn on unlimited-token-CEO-mode and ask Claude to devise a plan to out maneuver the military and intelligence services? It’s not AGI yet, and this request would be far outside the training distribution: it would just hallucinate something based on Tom Clancy novels.

Edit: typo

From what I read here, the required chip size would scale linearly with the number of model weights. That alone puts a ceiling on the size of model.

Also the defect rate grows as the chip grows. It seems like there might be room for innovation in fault tolerance here, compared to a CPU where a randomly flipped bit can be catastrophic.