oblio

I'll plug your comment into a couple LLMs. If it's so easy, they should be able to provide the numbers.

Edit: Gemini 3.5 Pro and Opus Claude 4.8 both disagreed that it's easy to determine anything, for both the high end (Fable, Sol) or the low end (open weights). Due to competition, subsidies, etc, gross margins could be as high as 85% (extremely unlikely) to as low as 10% or even negative. And that's just for pure inference and gross margins. Even for pure inference providers this doesn't include any overhead such as rent for office space for the pesky humans operating the business, marketing expenses, etc, etc. Let alone any crazy soul that actually wants or needs to train something.

Surprise: the comment above was downvoted in the bastion of libertarianism :-)

https://youtu.be/lh2__MN-FTU?si=LXIaljh__s8fD75l&t=1568

About 3 minutes of video worth watching.

Because Amazon needs to justify $175bn of yearly capex spending and $2.5tn of market cap? Amazon owns a big chunk of Anthropic and a bit of OpenAI.

For Magnificent 7 the AI bubble bursting will probably wipe out 30-50% of their valuations until the next tech cycle begins.

Libertarianism is not a philosophy. It's selfishness taken to extremes and trying to find ways to justify it at a societal level. The only reason we're the top species is because we're ultra social and have culture, which is inherently a social trait (don't eat those red berries, they're poisonous). Libertarianism want all the benefits of working together with no actual thought into how that working together happens in real life, including punishment for bad behavior.

Soon you will encounter LLMs on the information super highway!

and make inference cheap enough to eventually escape the red numbers

Besides training, we have no hard, externally audited numbers that say inference costs for SOTA models are truly sustainable. Do any OpenRouter providers have publicly audited financial numbers ?

Reality is a fractal (Mandelbrot?) and this is where LLMs will help science. They will help us explore the fuzzy edges faster. LLMs probably won't find new "leaves", but they should be able to cover a lot of useful but "boring" fuzz.

"There are 2 hard problems in computer science: cache invalidation, naming things, and off-by-1 errors."

I wonder if we're a 5-10 years from running this on beefy consumer hardware, or more like 20+ years.

Interesting. I wonder if this could work in the opposite direction, forcefully opening Apple hardware to other OSes. Think not just Asahi Linux, but regular Linux distros on Apple hardware.

Interesting, it reminds me a bit of the Linux in the browser thing.

I said a lot more than that and this entire discussion is in the context of high speed trains and light pollution.

Again, reading comprehension, context, and I guess, typical techie obtuseness.

Did you read my previous comment in detail?

People don't live in caves, they will create light pollution everywhere, anyway. Might as well put them in small and dense population centers.

I hate reading comprehension these days :-)

They can only do that if AI is profitable or at break even. Otherwise they would need to finance these companies yearly.

And... the people? Where do they come from? Mars?

All you're describing is population redistribution, not population creation. Urban centers have very low fertility rates, which actually reduces populations...

I assume ROM would be even more expensive than SRAM?

declined faster after 2019

Is your study discovering the effects of... COVID-19? I guess you've just discovered your own analogy of https://xkcd.com/1138/

the real problem is they can't rule that parents must be responsible for their children

It's impossible for the average parent to police their child's smartphone. They just don't have, and never will have, the technical skills.

I find the moral panic about kids with phones/internet access fascinating

And for those parents, which are quite rare, that do have the technical skills AND the involvement to ban smartphone usage at young ages, guess what. When Joey and Mickey and Sarah all have smartphones and are doom scrolling TikTok, even those parents lose. If all socializing moves to smartphones, the weirdo kids that don't have them or access to the "cool stuff" are automatically excluded. It's extremely alienating and nothing like not having calculators or PCs.

Back in your day (or mine), huge corporations more akin to drug dealers employing tens of psychologists weren't trying to get kids addicted to calculators.

but if there aren't population centers, you'll have to have plans to create them in order to support the rail system itself.

This is such a misanthropic point of view.

Nobody "creates" population centers. People aren't created out of raw clay, they already exist.

If population centers are "created" it's because people already exist and want to live in a specific place, often time urban.

And if they wouldn't go to these population centers they wouldn't live in caves in the dark. They would be spread over many more places, yes, creating lower levels of light pollution but lots more of them, covering a lot more land.

I've talked to parents of kids where they banned smartphones and the difference is palpable. And good.

This type of repetitive comment[1] is a bit like the bike lanes discussion:

1.City introduces bike lanes on street.

2. Residents complain about parking spaces going away, businesses complain about future bankruptcies due to fewer customers, general uproar is huge.

3. Residents discover bike lanes are quiet, and they also reduce ambient noise levels due to reduced car traffic, air is also cleaner, roads and intersections are safer.

4. Businesses (re) discover that cars don't buy things, people in big metal boxes going 50+kmph don't buy as many things. But people biking and walking are... foot traffic... and business goes up.

5. City decides to introduce bike lanes on another street.

6. Cycle repeats.

To your point about phones, it is very effective. If nobody can take their phone out on the school premise, suddenly it's very obvious who's trying to break the rule and guess what. Kids play, kids talk face to face, etc.

* * *

[1] This type of comment is one of the first things posted for these types of articles.

It's super hard to time the market. I would say the sanest approach is to have cash on hand, choose AI adjacent companies with solid fundamentals and buy their shares once the bubble pops. It's very hard to make money being an outright bear, but somewhat easier if you're an early phase/post recovery bull.

Outside of Google, has semantic search succeeded anywhere else?

Why haven't Anthropic and OpenAI produced a regular series of case studies where Company X did Fun AI Thing and incredible success ensued?

Even better, why aren't AI labs guaranteeing success if a certain formula is followed and asking for perpetual royalties for successful projects?

If one has a deep, conceptual understanding of tools, systems and protocols and if one has the ability to carefully provide careful, concise specifications

Most people already don't have the ability to do this and their use of LLMs atrophies the little they had. "AI natives" will be completely doomed.

1. Everything you said is more about revolutionizing IT though. IT is more easily testable than other things in real life.

2. The gap between the singularity and really strong automation is huge. It's basically that between 1 million and infinity, or even better, that joke about the difference between 1 million and 1 billion being roughly 1 billion.

municipal governments adversely affect traffic flow by lowering speed limits because those lower speed limits generate more ticket revenue.

I don't know about the US because the US is weird, but:

* at 30kmph the rate of fatalities in case of a car hitting a pedestrian is basically 0%, at 50kmph I think it's 5% or more

* at 30kmph collisions are much easier to avoid due to the increased reaction time and the decreased braking distance (I don't remember the exact numbers)

* at 30kmph you can hold a conversation at normal speech levels next to a moderately busy road, at 50kmph you will have to shout (and not even notice it due to the high ambient noise)

Etc.

Spoken as a German or as an American?

Tesla car sales peaked in 2023 and have declined since. They've retired the models S, X, and the CyberCab is barely selling.

Tesla also peaked at about 15% of Toyota sales. Tesla profit margins are now comparable to regular car manufacturers. Robots are a pipe dream. Robotaxis are a decade into the future as a meaningful business.

How exactly does Tesla's market cap make any sense?

Tesla will fall once they start going more into the red when:

* their products are recalled (just as for "legacy" manufacturers, because making cars is hard)

* Chinese brands take over every market where they're not outright banned

* FSD doesn't materialize because it's a decades long project that requires more types of hardware than just simple cameras

* robotaxis can't be rolled out because of no FSD and no approvals

* the robots go nowhere for decades

Basically, it turns out that despite the desperation, Tesla will slowly turn into a boring car manufacturer with just 2 car models for sale.