If someone uses an LLM as a replacement for search, and the output they get is correct, this is just by chance. Furthermore, a system that is right 95% of the time is arguably more dangerous tthan one that is right 50% of the time. People will be more likely to trust the output, and likely less able to fact check the 5%.
Logical properties, container queries, :has, :is, :where, min(), max(), clamp(), nesting, cascade layers, subgrid, and more.
I enjoyed reading through these essays about the web of twenty years ago: music, photos, email, games, television, iPods, phones…
Much as I love the art direction, you’d never know that we actually had some very nice-looking websites back in 2004!
This is a neat project form Dries:
This project is driven by my curiosity about making websites and web hosting more environmentally friendly, even on a small scale. It’s also a chance to explore a local-first approach: to show that hosting a personal website on your own internet connection at home can often be enough for small sites. This aligns with my commitment to both the Open Web and the IndieWeb.
At its heart, this project is about learning and contributing to a conversation on a greener, local-first future for the web.
This short essay by Richard Feynman is quite a dose of perspective on a Monday morning
A solid detailed in-depth report.
The sheer amount of resources needed to support the current and forecast demand from AI is colossal and unprecedented.
Obviously I’m biased, but I very much agree with Sophie.
I hope to make something that could only exist because I made it. Something that is the one thing that it is. Not an average sentence. Not a visual approximation of other people’s work. Not a stolen concept that boils lakes and uses more electricity than anything in my household.
At this point, it really does seem like “AI” is “bullshit you don’t need or is done better in other ways, but we’ve just spent literally billions on this so we really need you to use it, even though it’s nowhere as good as what we were already doing,” and everything else is just unsexy functionality that makes what you do marginally easier or better. I’m sorry we live in a world where enshittification is being marketed as The Hot And Sexy Thing, but just because we’re in that world, doesn’t mean you have to accept it.
This observation seems intuitively obvious in Europe and pearl-clutchingly shocking in America:
What’s perfectly acceptable behaviour when you are a relatively small company becomes outright illegal (and rightly so) when you become dominant in an industry.
“AI” is heralded (by those who claim it to replace workers as well as those that argue for it as a mere tool) as a thing to drop into your workflows to create whatever gains promised. It’s magic in the literal sense. You learn a few spells/prompts and your problems go poof. But that was already bullshit when we talked about introducing other digital tools into our workflows.
And we’ve been doing this for decades now, with every new technology we spend a lot of money to get a lot of bloody noses for way too little outcome. Because we keep not looking at actual, real problems in front of us – that the people affected by them probably can tell you at least a significant part of the solution to. No we want a magic tool to make the problem disappear. Which is a significantly different thing than solving it.
Its proponents can be weird, it takes itself far too seriously, and its documentation is interminable. These are some ways that some people have described Christianity. This video is about React.js.
Our ethical struggle with generative models derives in part from the fact that we…sort of can’t have them ethically, right now, to be honest. We have known how to build models like this for a long time, but we did not have the necessary volume of parseable data available until recently—and even then, to get it, companies have to plunder the internet. Sitting around and waiting for consent from all the parties that wrote on the internet over the past thirty years probably didn’t even cross Sam Altman’s mind.
On the environmental front, fans of generative model technology insist that eventually we’ll possess sufficiently efficient compute power to train and run these models without the massive carbon footprint. That is not the case at the moment, and we don’t have a concrete timeline for it. Again, wait around for a thing we don’t have yet doesn’t appeal to investors or executives.
Using ChatGPT to complete assignments is like bringing a forklift into the weight room; you will never improve your cognitive fitness that way.
Another great piece by Ted Chiang!
The companies promoting generative-A.I. programs claim that they will unleash creativity. In essence, they are saying that art can be all inspiration and no perspiration—but these things cannot be easily separated. I’m not saying that art has to involve tedium. What I’m saying is that art requires making choices at every scale; the countless small-scale choices made during implementation are just as important to the final product as the few large-scale choices made during the conception.
This bit reminded me of Simon’s rule:
Let me offer another generalization: any writing that deserves your attention as a reader is the result of effort expended by the person who wrote it. Effort during the writing process doesn’t guarantee the end product is worth reading, but worthwhile work cannot be made without it. The type of attention you pay when reading a personal e-mail is different from the type you pay when reading a business report, but in both cases it is only warranted when the writer put some thought into it.
Simon also makes an appearance here:
The programmer Simon Willison has described the training for large language models as “money laundering for copyrighted data,” which I find a useful way to think about the appeal of generative-A.I. programs: they let you engage in something like plagiarism, but there’s no guilt associated with it because it’s not clear even to you that you’re copying.
I could quote the whole thing, but I’ll stop with this one:
The task that generative A.I. has been most successful at is lowering our expectations, both of the things we read and of ourselves when we write anything for others to read. It is a fundamentally dehumanizing technology because it treats us as less than what we are: creators and apprehenders of meaning. It reduces the amount of intention in the world.
Very good writing advice from Dan:
Use plain language. Tell the truth.
Related:
The reason why LLM text for me is bad is that it’s insipid, which is not a plain language word to use, but the secret is to use words like that tactically and sparingly to great effect.
They don’t write plainly because most of the text they’ve been trained on isn’t plain and clear. I’d argue that most of the text that’s ever existed isn’t plain and clear anyway.
A library of CC-licensed photos.
Next time you’re tempted to use a generative “AI” tool to make an image for a slide deck, use this instead.
The human desire to connect with others is very profound, and the desire of technology companies to interject themselves even more into that desire—either by communicating on behalf of humans, or by pretending to be human—works in the opposite direction. These technologies don’t seem to be encouraging connection as much as commoditizing it.
Benjamín Labatut draws a line from the Vedas to George Boole and Claude Shannon onward to Geoffrey Hinton and Frank Herbert’s Butlerian Jihad.
In the coming years, as people armed with AI continue making the world faster, stranger, and more chaotic, we should do all we can to prevent these systems from giving more and more power to the few who can build them.
Number one:
Do things in the most straightforward way possible. It’s easy to fall into the trap of clever solutions, or clever applications of technology, or overbuilding something because you’re anticipating the future. Don’t do it. You will hate yourself for it later when you have to maintain it.