csomar

> Datacenters have massive economies of scale.

Sure. Issue is, no one is providing on how much it actually costs to burn these tokens. And as we don't know, we can only speculate.

> Many say that, but I sincerely doubt they'd actually follow through.

I have a $100 open ai sub and I track my token usage. Last month I spent roughly $2.600 in equivalent API usage. There is no way am paying that. I let my $100 sub lapse if next month I'll be using it less.

Look, I am not saying that there isn't a potential value out there. But the cost has to be bounded. If your opportunity is $1.000 and AI costs $2.000 to execute it, then you don't have a business model here.

I think it's sinister, but not for the reasons you're thinking. I think they're just wildly unprofitable on subscriptions. The idea that most customers won't use their full quota is plain wrong: most people are maxing out their subs, or even reselling whatever quota they have left.

When you're running something at a loss, you can mistreat your customers and they'll still stick around (I'm an example). OpenAI and Anthropic are now cheaper than Chinese models on subscriptions, while being 6-10x more expensive on the API.

My guess is they need the user numbers for the IPO and are willing to take a temporary loss in the meantime. By the time they go public, they'll either drop the subscription model or it'll turn into what the Chinese providers already offer: basically just a cap on how much API you can consume. Same same.

It's not clear what API tokens actually cost them, but I looked into running a local model, and it's way outside the budget of an individual or even a small or medium business (hundreds of thousands of dollars). So my guess is that running these models economically isn't possible, even if they're delivering real business value (coding, research, etc.). In other words, at API prices I'd just stop using AI, and I suspect most other developers would too.

They would have recharged the reserves if that was really the reason and they would have readied themselves to the upcoming closure. They didn't expect the strait to be closed and got caught with their pants way down. Using the SPR is just another sign of how incompetent and out of options they are. It was already on a low level when the war started due to the previous withdrawals from the Russian war.

When Microsoft did that there wasn't really any competition and they were also the cheap option. I don't think they would survive if there was a sufficiently good chinese OS at the time.

CEOs only job is to raise the stock price. They are doing just that and right now the way to get people to buy into their stock is to instill fear into them that they need to buy the AI-stock for their safety. There is no "bigger conspiracy" out there and also no AI to kill us all tomorrow.

This has been that way in Silicon Valley since the very beginning and there was even a documentary about (the SV HBO series). The tech made along the way is just a side-product.

They are probably caching the answers. AI-answers are most likely never going to be profitable even with a smaller model.

The author probably has a very superficial understanding of economics. It's economics 101, like VAT, the cost will be passed to BOTH the renter and the landlord. The portion of each is dependent on market dynamics and hard to calculate.

It's the same language of GLM thinking tokens. My guess is that Opus 5 doesn't talk like that because we don't see the thinking tokens in Claude?

I mean we are in the thread of evidence right now?

I find it interesting that China today is moving in the opposite direction and going as far as banning low-quality exports and stopping this “churn” of products having learnt that low-value high-quantity products are a bad deal all around.

Given how sloppy AI without human directions, I’d like to see evidence that this was not human-directed. Against the prevalent opinion here, I’d give openai a pass if this was really fully autonomous ai agents.

My money is on special teams co-ordinating these agents and exposing their traces in order to create a pre-ipo buzz. Sounds ridiculous and reckless? Well that’s the AI industry for you in two words.

The issue is that this is the thing that got “noticed”. The problem is that your program behavior is now unpredictable. Anything can happen!

Apparently it takes 737 min. for the US to descend and become just a less-frosty USSR. What a time to be alive. I was born after the fall of the USSR so I didn't get to watch that but here I am for a bigger show.

Now label me as an adversarial agent to the great nation or given my ethnicity the "terrorist" label will be more befiting.

And not using authentication.

> Peer authentication in the connection handshake is broken and allows impersonation. An attacker can connect to your node and present a Node ID that is not its own. Private repositories are shared only with allow-listed Node IDs. An attacker who fakes an allow-listed Node ID can fetch a private repository directly, without being on the network path. This was reported to us by cryptocode on 2026-08-12. We proposed a fix upstream, see this pull request.

They are trying to sweet write it as much as possible. But basically there is neither encryption nor authentication. The person who made the protocol/program simply didn't care.

Does it use less tokens or we just get no accounting of the thinking tokens in OpenAI/Claude models?

My experience with Rust Wasm workers was that they eventually had shorter cold-start and runtime than JavaScript ones because your wasm is significantly more optimized than a JavaScript one that loads whatever packages/dependencies into its memory.

> and doesn't have second thoughts.

and so is the Trump administration?

Instead of funding a skin UI for Linux, that money could have been used better to fund Asahi Linux.

Many apps in the App Store are now 0.5Gb in size. I got an M5 with 2 Terabytes a couple weeks ago and I was wondering why do I need all this space for. I guess I might end up short on space.

Most people who comment positively on AI are either shills or people working in the AI industry (gotta keep the bubble expanding). The OP experience matches mine with most models for any kind of serious work. In fact, it was the reason why Claude was previously successful because it was the most railed model.

I've tried the latest models across OpenAI, Claude, Chinese, etc. They just do stuff. That's not how work is though. You want them to do specific work, at which point it's a real hassle to follow up on all the garbage they have been outputting.

/long rant

I thought this would be about the non-fungibility of money and was greatly disappointed. Some (many? most?) people are not aware that a dollar here is not equal to a dollar there and this got way worse in the last few years. It has become exceedingly difficult to move value around even in the same country (despite the hundredth fintech starting up to solve this problem and move money between friends)

The last guy (kimdotcom) who was working on this (incentive for seeding) seems to be heading to the US: https://www.rnz.co.nz/news/science-and-technology/651123/cou...

It’s interesting he’s no longer getting any media attention any more.

How does this address the previous point? If they are not providing storage, then centralized or decentralized doesn’t make a difference.

Torrent/P2P can only add redundancy, so it’s impossible to have worse availability than a download link?

Before anyone even contemplates if Anthropic has any good intentions, remember, this is how petty and small minded they are.

Truly the last people you want with this kind of power.

The issue (real issue?) is that it’s unclear whether these governments reduced their holdings or switched them to these opaque structures (tether can be considered one). The idea is, it would be hard for the US to untangle true ownership. I wonder if UBO was getting undone blue or red because it’s a real threat for such a system but the US needs this “second” lifeline.

Based on historical and current events (USA) it takes roughly a couple of years for a country and its leaders to flip from pacifist to aggressive.

We’ll have the same level of security as before; it’s just that, without LLM help, hackers won’t be as effective as before. So the bar is raised.

That's really funny when you consider that generative models also don't hallucinate if you check up on them on every token generated?

If you allow the law to stop you from doing something because of a hypothetical danger, it’ll get severely abused. I come from a country that does and still does this and it’s very ugly.

I am surprised how no one is mentioning how their sneakers dissolves in a couple months? Last store Nike sneaker I bought was 4 years ago. $120 or something and within 2 months it has become unwearable. Since then I only buy new balance from an outlet so the price remains below $100.

Still the last new balance I got are wearing out fast and I am on the look for a newer brand.

I am actually surprised they are still in business and making sales.