I’m quite happy with my electric razor though ;) But yeah, single-use plastic products, and their implicit “subscription”-like business model, is a good analogy.

I also don’t expect that Gen-“AI” will go away entirely anymore, it’s too useful for generating low-quality crap for e.g. spam and disinformation and similar purposes. I also dread the thought that when I buy a translated book now, I won’t know much it was actually translated by a person.

However, I still have hope left that it will eventually become more of a background noise. Like how cryptocurrency still exists now, but at least we don’t have to hear anymore about how “NFTs are the future of art” (just remember what a common theme that was for a while, pretty recently). Likewise I think that “AI is the future of <creative thing>” will eventually fade away.

And some people are already creating and spreading little “human made” seals that one can attach to projects, I hope that catches on, like labeling of food products. And not just in niches like open source software (where I’ve seen it so far), but widely across all kinds of creative things, like book translations and music and so on. I can hope, right?

Once the big hype is over, when the bubble has burst, the absolutely enormous costs of running all the server farms will have to be passed on to the product-making companies and they will have to further pass it on to their users. As a result I think that most of these “AI” “features” will be pulled from most products, because who’s really willing to pay for it? And I don’t expect that it will become cheap soon. In their desperate attempts to make their “AI” perform “better”, the companies are currently cranking up the usage of compute power to ever-higher degree, because they’re otherwise out of ideas how to improve anything about it. And from what I hear (out of principle I never use this stuff myself), the small models which one could run locally just aren’t very good (not that the big ones are “good”…). (However, as written above, they will always be good enough for spam/disinformation and such where quality doesn’t matter.)

So I don’t believe that this will be like the 80s or 90s, where one could develop fancy big software with the expectation that within a few years even the cheap entry-level machines will be fast enough for it. That kind of performance progress stopped long ago. I expect that this will stay really expensive for the foreseeable future, at least for the “better” models. And then, maybe, with most of this crap pulled out of our tools for plain and simple reasons of “cost”, together with the collapse of the hype around it, we can go back to this being mostly background noise.

Yeah, I’ve always been kind of a hopeless optimist…

On the (slim) upside, it’s an opportunity to ditch Google, and maybe it will sooner or later break their monopoly position. I switched my main search engine to Ecosia a while ago, I think it uses Bing underneath (meh), but presumably it’s more privacy friendly than Google (or Bing directly). I’ve had numerous such attempts over the years already to get away from Google, but always returned, because the search results were just so much better (especially for non-English stuff). But now Google has gotten so much worse that it created almost an equilibrium… sometimes it’s still useful and better, but not that often anymore. So I rarely go to Google now, not because the others got better, but because Google got so much worse.

(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.