nightsky @ nightsky @awful.systems

sometimes a dragon

---

Read more

Posts

0

Comments

167

Joined

11 mo. ago

(found here:) O'Reilly is going to publish a book "Vibe Coding: The Future of Programming"

In the past, they have published some of my favourite computer/programming books... but right now, my respect for them is in free fall.

There's an article about it at 404 media now. A few quotes from the founder guy:

"[...] I think it would be foolish to think that there is a future where ChatGPT doesn’t come in your glasses."

“We want to make everybody use AI. We are true AI maximalists.”

“Entire school systems will be gone. The entire K-12 education. [...]"

And he gets millions of investment... just deranged late-stage bubble stuff. It's sad to think how much good $5.3 million could do, e.g. for actual education, and instead it's blasted away for creating such garbage.

It really sucks so much how many coders embrace it. At my work, there is the looming introduction of code LLMs very soon, and I'm anxious to learn how many of my colleagues will happily use it, and the consequences it will have for me to deal with the results (and generally, how it will make me feel to work in an environment where these tools are embraced). I was hoping that the corporate bureaucracy would be slow enough that the AI bubble collapses before it's allowed to use the tools, but unfortunately management put a lot of pressure behind it and it all went faster than expected :(

404 media: I Tested The AI That Calls Your Elderly Parents If You Can't Be Bothered

It's a service that makes an AI voice chatbot call your parents daily, so you don't have to, and then it even sends you a notification to your phone with an AI summary of what your parent told the AI.

I really didn't think that people can come up with new AI-based ideas anymore that would astonish me, but there, I was wrong, they did it. This is so cold and fundamentally alienating to me, it reminds me of that recently much-quoted Miyazaki phrase, "an insult to life itself".

I hope that either way the opportunity will be taken to move this all to a safer/independent footing.

No more CVEs, so I guess that means no more vulnerabilities, the computer security crisis is solved, who knew it would be that easy!

It's a complete shitshow and very scary, even just looking at it from the outside, can't imagine what it must feel like from the inside. I keep having to remind myself that all these things that currently happen are real.

Yes, that's true. Indirectly it costs them all dearly with ransomware. Likewise, I think the overall damage that AI will do to society as a whole will be much, much greater than just rotting some tech companies from the inside (most of which I wouldn't be sad anyway if they went away...).

What I meant is that with blockchain the big tech companies at least didn't willingly destroy their products, their processes, their decision making etc. I.e. they didn't put blockchain into absolutely everything, all the way to MS Notepad. What I find staggering about this hype is the depth of the delusion, the willingness to not just experiment with it but really go all-in.

That text is painful to read (I wonder how much of it is slop)... ugh, what is chatgpt doing to the brains of people? (And I've had the bad luck of reading some pretty unhinged pro-AI stuff from management at my employer too, although not as bad as this mail from shopify).

Is there a precedent for this hype? For the extent of damage that it will cause? Most tech industry hype is a waste of resources, but otherwise mostly harmless. Like that time when everyone believed that XML is the holy grail, that was silly, and although we still have to deal with some unfortunate data formats from those days, it passed. There were worse ones, most notably blockchain was almost catastrophic, but most companies hesitated to go all-in and pursued it more on the side, so when that hype faded, they simply buried their involvement and that was that.

But "AI"... it has such potential to create significant and long term damage to the companies adopting it. The slop code alone might haunt them forever, in ways that even the worst excesses of 90s enterprise java couldn't. There's nothing to learn from resulting failure, except "don't use AI".

In this case, given shopify's general behaviour, I won't be sad at all though if they crash and fail.

60 million lines of COBOL code today and millions more lines of Assembler

Now I wonder, is this a) the most extreme case of "young developer hybris" ever seen, or b) they don't actually plan to implement the existing functionality anyway because they want to drastically cut who gets money, or c) lol whatever, Elon said so.

But no no, surely they just need the right prompt. Maybe something like this: [...]

Labrador retrievers ;_; You're getting too good at this...

sam altman is greentexting in 2025

Ugh. Now I wonder, does he have an actual background as an insufferable imageboard edgelord or is he just trying to appear as one because he thinks that's cool?

That's much better!

Damn, I should also enrich all my future writing with a few paragraphs of special exceptions and instructions for AI agents, extraterrestrials, time travelers, compilers of future versions of the C++ standard, horses, Boltzmann brains, and of course ghosts (if and only if they are good-hearted, although being slightly mischievous is allowed).

The article already starts great with that picture, labeled:

An artist's illustration of a deceptive AI.

what

Good one!

On the other hand, your book gains value by being published in 2021, i.e. before ChatGPT. Is there already a nice term for "this was published before the slop flood gates opened"? There should be.

(I was recently looking for a cookbook, and intentionally avoided books published in the last few years because of this. I figured that the genre is a too easy target for AI slop. But that not even Springer is safe anymore is indeed very disappointing.)

A day later and I'm still in disbelief about that windsurf prompt. To make a point about AI, I think in the future you could just show them that prompt (maybe have it ready on a laminated card) and ask for a general comment.

Although... depending on how true the true belief is, it might not have the intended effect.

Trying to imagine the person writing that prompt. There must have been a moment where they looked away from the screen, stared into the distance, and asked themselves "the fuck am I doing here?"... right?

And I thought Apple's prompt with "do no hallucinate" was peak ridiculous... but now this, beating it by a wide margin. How can anyone claim that this is even a remotely serious technology. How deeply in tunnel vision mode must they be to continue down this path. I just cannot comprehend.

Whatever has happened there, I hope it will resolve in positive ways for her. Her amazing work on the GPU driver was actually the reason I got into Rust. In 2022 I stumbled across this twitter thread from her and it inspired me to learn Rust -- and then it ended up becoming my favourite language, my refuge from C++. Of course I already knew about Rust beforehand, but I had dismissed it, I (wrongly) thought that it's too similar to C++, and I wanted away from that... That twitter thread made me reconsider and take a closer look. So thankful for that.

Reuters: Quantum computing, AI stocks rise as Nvidia kicks off annual conference.

Some nice quotes in there.

Investors will focus on CEO Jensen Huang's keynote on Tuesday to assess the latest developments in the AI and chip sectors,

Yes, that is sensible, Huang is very impartial on this topic.

"They call this the 'Woodstock' of AI,"

Meaning, they're all on drugs?

"To get the AI space excited again, they have to go a little off script from what we're expecting,"

Oh! Interesting how this implies the space is not "excited" anymore... I thought it's all constant breakthroughs at exponentially increasing rates! Oh, it isn't? Too bad, but I'm sure nVidia will just pull an endless amounts of bunnies out of a hat!