techpression

> We’ve installed our own servers, co-located in a secure facility in Amsterdam, set up by our own engineers. This new location is built to the same high standards as our existing infrastructure in Philadelphia and St Louis, with our own hardware and our own software — specified right down to the exact model of disks in each machine.

I think most people think the ethical values are far higher in those jobs than most random startup bs just wasting resources.

I definitely do, I would rather work in the military than Meta/Amazon/Microsoft/Any online gambling/and loads of other shit companies.

People seem to think LLMs are sentient and not tightly coupled to server infrastructure OAI are in full control over. They literally have every single message sent and action taken, if they can’t monitor that during test runs they should find another job.

It’s like running a benchmark and forgetting to tune it and then wondering why production is running slow.

You also need to worry about yields, Apple, AMD etc can sell ”bad” chips as lower core versions, if you’re depending on whole wafer you have little room for error.

I didn’t get far until the semantics fall apart, you don’t care if it’s consumption or not when you don’t have water in your tap. And it’s all based on theory, these plants are using systems far under dimensioned for the task, which is never a good thing. The problem with proponents for these behemoths is that they are all about theory and don’t live anywhere near the life they say is so great.

https://youtu.be/2Eltf06-NiQ

Feels like you’re cherry picking to make a point, so here’s some cherry picking on the other side https://www.fierce-network.com/data-center/top-10-ugliest-da...

So much dead ugly space, and let’s not look at them at night with enough external lighting to see from space (almost)

What? How is a data center that runs insane amounts of water through it every single day the same as a warehouse with a tap and bathroom?

And sure, a rocket launch pad would also cause issues, but now it’s a data center that is doing it, let’s not argue fictional realities.

One piece of load-bearing coming right up

Still waiting for them to fix the limitations of their Google tag gateway, but considering their focus over the last year I doubt it will happen.

Just like phishing websites, which are still a problem. I’m sure I’m being overly cautious, but this sounds like a recipe for data leaks with the low barrier to entry.

And if you want real life examples, every single LLM provider has gotten insane amounts of data shared to them, even if you have to approve.

But that is a problem. I have access to something deeply secret, I run a gadget, it reads that and sends it to whoever made it

This is what most people seem to forget, OpenAI has probably spent more money the last year than all of math research has during human existence.

Do you verify the reply on how Z works too?

> The same thing happened with their massive investment in Anthropic. They accounted for $53.4 billion due to deals with Anthropic last quarter. He said if you follow one Anthropic dollar through the earnings release, it's counted in AI business revenue, chips business, and AWS segment revenue.

That is insane if that is true, is that even legal?

It probably has a few million inputs on how a pelican and bicycle looks, not to mention the amount of data on how to create SVG’s. Ask it to create a relaxing spa website and it will, even though it has never seen a spa.

When is it ever hard to catch?

There’s an infinite number of ”working stuff” implementations, which is kind of the problem, bad decisions multiply. That said, looking at small parts of code for improvements tend to work very well for me, optimizing things like data structures, iterations and SQL queries.

I just hope my bank doesn’t go Claude code all the way any time soon.

This mirrors my experience too, long term use starts showing incoherence that is impossible to see in isolation.

The funny thing with LLM’s is that with a sufficient sized code base they pretty much will loop forever if you tell them to find and fix issues, making new ones as they fix old ones.

Everyone can drive really fast if you give them a fast car, but very few can stay on the road when driving fast. LLMs are like free supercars for everyone, except people know nothing about how supercars work or how to drive them. Same thing with software, going fast is incredibly hard if you do it for any period of time, and I think this is something we will read more about in a couple of years, not that human written code never turned out to be complete disasters, but I believe with LLMs we will see at a completely different scale. And yes, car analogies suck.

I spent 30min with Opus 5 generating some CSS and components for my Phoenix LiveView template project (which I use for starting new projects). Then I spent around 5 hours parsing and cleaning up the output. I could of course not have done that, but since it's my spare time project I can care about the code and quality.

I did get wins for sure (and I did save time), and I'm sure most people wouldn't even spend five minutes cleaning up, but this is definitely one place where people talk about different things regarding whatever x speedup they get.

Agree, not much special about them, and I truly hate their keycap profile which I find to be way too tall. They’re not bad keyboards, and regularly on sales in the 40-60% range for many of their boards, it’s just that they’re not great either, middle of the road so to say.

Chinese labs are equivalent to the Chinese state, there’s no separation. Who do you think funds the Chinese labs?

It’s not a service, I hope nobody thinks this is being done ”for the good of humanity” or whatever. Doesn’t mean there’s not a benefit to people, but the financial politics behind it could turn out extremely good for China, and they are very well aware of just that.

Whatever happened with their new programming language for agents? I expect this will follow the same path.

Pretty insane that we had both Ericsson and Nokia making really good phones and then getting left in the dust. As for GrapheneOS it doesn't sit well with the desire for total digital oversight from EU, digital ID's, tracking and invasive things like on-device scanning, unfortunately.

Where’s the data for one, do they include the force fed Gemini in search? They are also pushing it harder than Microsoft was pushing copilot in every single of their products, sometimes multiple competing implementations (hello GCP dashboards).

It simply comes off as a marketing advert for a company who’s sunk billions upon billions into something they’ve yet to be able to monetize (see latest report). Stock holders are getting annoyed, like in other companies, about the lack of the promised returns, this is firefighting.

Nobody is really debating AI has a use case, but maybe, just maybe, the chase into infinite spending we’ve seen the last years wasn’t that well thought out.

Apparently it’s already happening to a degree, wether it continues or not (or even is relevant) is not really my area of expertise.

https://finance.yahoo.com/markets/stocks/articles/goldman-sa...

The Hacker part of Hacker News, thanks for sharing! I think you'd get a lot of bowling alley friends all over if you can package this in a way that they can use it, I would assume the Venn diagram of people running bowling alleys and those who know what ESP32 and RS485 means is pretty thin.

Recursion is such a beautiful thing

Printing it on paper and manually looking it up is probably cheaper with today’s ram-prices