jonas21

I have no idea what you are even asking Claude to do here.

If you click through to the study that the Guardian based this article on [1], it looks like it was done by an SEO firm, by a Content Marketing Manager. Kind of ironic, given that it's about the quality of cited sources.

[1] https://seranking.com/blog/health-ai-overviews-youtube-vs-me...

It does have these details for the current generation hardware. And if you want more, click on the link at the top:

https://hackernoon.com/the-long-now-of-the-web-inside-the-in...

There's also an entire section on "what constitutes genuine helpfulness"

AI capex may or may not flatten in the near future (and I don't necessarily see a reason why it would). But smartphone capex already has.

Like smartphones, AI chips also have a replacement cycle. AI chips depreciate quickly -- not because the old ones go bad, but because the new ones are so much better in performance and efficiency than the previous generation. While smartphones aren't making huge leaps every year like they used to, AI chips still are -- meaning there's a stronger incentive to upgrade every cycle for these chips than smartphone processors.

Yes, that's my point. Neither driving nor coding with an LLM is perfectly deterministic. You have to learn to deal with different things happening if you want do do either successfully.

Operating a car (i.e. driving) is certainly not deterministic. Even if you take the same route over and over, you never know exactly what other drivers or pedestrians are going to do, or whether there will be unexpected road conditions, construction, inclement weather, etc. But through experience, you build up intuition and rules of thumb that allow you to drive safely, even in the face of uncertainty.

It's the same programming with LLMs. Through experience, you build up intuition and rules of thumb that allow you to get good results, even if you don't get exactly the same result every time.

There are plenty of non-blockchain, non-NFT, non-online gambling, non-adtech, non-facist software jobs. In fact, the vast majority of software jobs are. You can refuse to work with all of these things and not even notice a meaningful difference in career opportunities.

If you refuse to work with AI, however, you're already significantly limiting your opportunities. And at the pace things are going, you're probably going to find yourself constrained to a small niche sooner rather than later.

Yeah, it's apparently pulling in over $800K in annual revenue [1].

EDIT: Doing the math on the sponsor list, it's probably around $1M in ARR now.

[1] https://petersuhm.com/posts/2025/

> Electricity prices near data centers go up, right?

I hear this a lot, but the most comprehensive study I've seen found the opposite -- that retail electricity prices tend to decrease as load (from datacenters and other consumers) increases [1].

The places where electricity prices have increased the most since 2019 (California, Hawaii, and the Northeast) are not places where they're building a lot of new datacenters.

[1] https://www.sciencedirect.com/science/article/pii/S104061902...

I can generally get maintainable results simply by telling Claude "Please keep the code as simple as possible. I plan on extending this later so readability is critical."

The "obvious" solution that I think you're describing is the Hammersley sofa (1968), which has A=2.2074.

For a long time, it was thought this might be the optimal shape, but it was never proven. And it couldn't have been because it turns out that you can do better: the Gerver sofa (1992) is a more complicated shape, composed of 18 curve segments and has A=2.2195.

Nobody knew whether there might be an even better shape until now (assuming the proof holds up).

See: https://en.wikipedia.org/wiki/Moving_sofa_problem

You're getting at the crux of the issue: it's very hard to distinguish legitimate DMCA takedown requests submitted by individuals from illegitimate ones, and occasionally, they're going to make mistakes. Anything that the author said in his emails could have just as easily been said by someone else who was trying to take down the content illegitimately.

At the end of the day, the best option is to use an attorney who knows the right procedures and would also run the risk of professional consequences if they submitted false claims.

Seems like the sharp decline started shortly after they were sold to a private equity firm.

There's a minor issue with the calculations. It should be:

    60 * 445 / 216.276 = 123.453365145
    
    60 * 445 / 216.282 = 123.449940356

Not the other way around. And since the timing is only given with millisecond accuracy, the bpm should be rounded to the same number of significant digits:

    60 * 445 / 216.276 = 123.453
    
    60 * 445 / 216.282 = 123.450

So, it's the YouTube version that's 123.45 bpm to within the rounding error.

> Similarly, I don't understand why the CxOs in my current (BigTech) company still work. You're done. You can do anything you want

Has it occurred to you that perhaps what they want is to be the CxO of a big tech company? There’s a lot of power, prestige, and impact on society that you can’t easily have if you quit. Maybe they really enjoy the work itself too.

The title should be changed to "LLVM AI tool policy: human in the loop".

At the moment it's "We don't need more contributors who aren't programmers to contribute code," which is from a reply and isn't representative of the original post.

The HN guidelines say: please use the original title, unless it is misleading or linkbait; don't editorialize.

I’d suggest that the title should be, “Competition for Google AdWords is so strong that unsophisticated advertisers can no longer get a good return. Where do I promote my business now?”

The code is licensed [1] under the "Apple MIT" license [2], which is considered open-source. The weights are under a different, more restrictive license. This is mentioned at the bottom of the README.

[1] https://github.com/apple/ml-sharp/blob/main/LICENSE

[2] https://fedoraproject.org/wiki/Licensing/Apple_MIT_License

The cadence of the Claude Code team is pretty impressive. I count 57 releases since 2.0 (which was less than 3 months ago), and most of these had significant new features.

An alternative business model is that you, or more likely your insurance, pays $20/mo for unlimited access to a medical agent, built on top of an LLM, that can answer your questions. This is good for everyone -- the patient gets answers without waiting, the insurer gets cost savings, doctors have a less hectic schedule and get to spend more time on the interesting cases, and the company providing the service gets paid for doing a good job -- and would have a strong incentive to drive hallucination rate down to zero (or at least lower than the average physician's).

You could make the same argument that "Using Open-Source Code Will Make You a Bad Programmer" -- and in fact, a generation ago, many people did.

Yeah, it's basically free publicity for them.

They have a major outage every couple of years. These don't have as much impact as some other cloud outages because they only affect Apple products, i.e. they're not hosting services for other companies.

- https://www.cnbc.com/2024/04/03/apple-investigates-app-store...

- https://www.macrumors.com/2023/02/23/app-store-apple-music-a...

- https://www.the-sun.com/tech/4944089/apple-maps-down-icloud-...

- https://www.macrumors.com/2019/05/08/itunes-and-app-stores-s...

- https://www.macrumors.com/2018/03/27/app-store-outage/

- https://www.cnbc.com/2016/06/02/apple-reporting-outages-for-...

- https://www.cnbc.com/2015/03/11/some-apple-services-sufferin...

That's what makes it fun. Apparently, Gemini has a better sense of humor than HN.

There was a response to this post on the front page earlier this morning that was able to get Claude to succeed simply by giving it access to Playwright so it could see what it was doing and telling it in the prompt that it needed to be pixel perfect:

https://news.ycombinator.com/item?id=46193412

As of right now, it seems to have been flagged into oblivion by the anti-AI crowd. I found both posts to be interesting, and it's unfortunate that one of them is missing from the conversation.

> their huge customers will be able to produce ASICs hat will be faster and cheaper to operate than their GPUs

Why? NVIDIA is better positioned to produce faster and more efficient ML ASICs any of their huge customers (except possibly Google). And on top of that, the fact that there is a huge library of CUDA code that will run out of the box on NVIDIA hardware is a big advantage.

Arguably, this shift has already happened. Modern NVIDIA datacenter GPUs, like the H100, only bear a passing resemblance to a GPU -- most of the silicon is dedicated to accelerating ML workloads.

> Where's the emdash key on your keyboard?

The dash key is right between the "0" and the "="

Press it twice and just about every word processing program in existence will turn it into an emdash.

Indeed, this line of reasoning goes all they way back to Socrates who argued that outsourcing your memory to writing would make you stupider [1]:

> For this invention will produce forgetfulness in the minds of those who learn to use it, because they will not practice their memory. Their trust in writing, produced by external characters which are no part of themselves, will discourage the use of their own memory within them. You have invented an elixir not of memory, but of reminding; and you offer your pupils the appearance of wisdom, not true wisdom, for they will read many things without instruction and will therefore seem to know many things, when they are for the most part ignorant and hard to get along with, since they are not wise, but only appear wise.

I, for one, am glad we have technologies -- like writing, the internet, Google, and LLMs -- that let us expand the limits of what our minds can do.

[1] https://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext...

Those are there because the content creator you’re watching decided to put them there. It’s entirely up to them whether they show up and when they show up.