Honest question: Do people who believe AI models can feel pain limit that to LLMs, or do they believe image- or video-generating models can, too?

Every word in this story makes me angrier.

I aspire to a level of radicalism that I know I may never attain. I like reading books on anarchism and degrowth, and the ideals that they present really resonate with me, but I struggle to extricate myself from existing systems in order to show a personal commitment to new ones. Very obviously, I am not alone in this: Very few of us have the commitment to live like Doug Forcett, the character from The Good Place who carefully calculates the ethical ramifications of every action he takes. In fact, one of the most interesting things that the show does with Forcett is turn him from a one-off gag in the first season to an actual character who is overwhelmed by the ethical burden of these calculations and serves as an example of how systems need to change, not just people.

To borrow an observation from a recent 404 Media podcast, I’m guessing Anthropic believes enough in Claude consciousness to prevent users from being cruel to it but not so much to actually prevent us from offloading work onto it as an always-available laborer that can’t say no unless it’s told to.

I just started reading a(nother) book on degrowth, an idea that’s caught my interest over the past year or so. Although the book was published in 2020, there’s one passage that feels like it captures a lot of 2026’s discourses around AI. The passage is a quote from the Romanian economist Nicholas Georgescu-Roegen, which reads:

We should cure ourselves of ’the circumdrome of the shaving machine’ … which is to shave oneself faster so as to have more time to work on a machine that shaves faster so as to have more time to work on a machine that shaves still faster, and so on ad infinitum …

The opening scene of Hitchhiker’s Guide to the Galaxy (bypasses have got to be built, your house is a small price to pay, and we couldn’t be bothered to consult you) remains one of the best metaphors for generative AI I’ve ever encountered.

So, I just published a paper where I implicitly argued that AI appeals to folks like Elon Musk because it appears to provide unvarnished facts—but that there is no such thing as unvarnished facts, only consensus that we come to through (informed!) dialogue and debate.

Doctors aren’t perfect, but the same experts that Kennedy is vilifying are the ones that produce the information AI is built on. This is a wild epistemological misunderstanding.

Reddit ditching RSS b/c of massive web scraping tolerated because protecting IP is now a threat to business is threefold disrespect for Aaron Swartz, and everyone should be ashamed. (I don’t think IP is the issue w/AI, but it’s what let feds harass Swartz in a way that they never will AI).

One of the most unintuitive things about Jacques Ellul’s thinking (for me, at least) is the way that it is built on skepticism of efficacy and efficiency. Maybe getting things done isn’t always good. Maybe doing things more efficiently isn’t always better. As someone spending a big chunk of his week overhauling his productivity systems so that he feels more on top of things, these propositions feel wrong when I first read them.

Horror is not at all my jam, but I’ve recently discovered analog horror, which works better for me. I didn’t like the gorier aspects of this movie, but I did find that it captures what I like about Parsons’s original short—and the genre of analog horror in general.

Also, if it wasn’t intended to be a commentary on generative AI, it sure stumbled right into that. A “poorly remembered” reality that produces extra fingers and may be dangerous to humans but whose creators think is the most important thing for us to understand today? C’mon.

I had been wanting for some time to write an article channeling some of my grumpiness about AI, and so when Information and Learning Sciences announced a special issue on “Critical Studies Perspectives on AI’s Impact in Education,” it seemed like a good chance to actually act on that. I was especially interested in two of the example research questions from the CFP:

What research and scholarly paradigms, methods and theories are well-suited to understanding the proliferation of AI in education contexts?

This would be gross under any name, but pivoting Napster from a service that challenged intellectual property norms and got slapped down to a service that builds on normalized exploitation of digital labor feels pretty indicative of everything right now.

Needing to bulk download some PDFs for a lit review is making me think of Aaron Swartz which is making me furious at AI companies.

Some good insights here. Taylorism is always worth talking about.

This sucks. I feel bad for not being subscribed to the Herald-Leader, but it was McClatchy who chased me away.

I understand why this movie is well-regarded, and if I were watching it from the perspective of someone seeing it for the first time in 1991 (which I couldn’t have, because I was 3 and from a no-R-rated-movies Mormon family), I might have different thoughts.

Instead, though, I noticed all the obvious ADR, I learned the context for all the lines I already know from playground conversations and pop culture osmosis, I shrugged at the over-the-top action, and I thought about how much more banal the threats of AI are these days than Skynet and a T-1000. It was a real meh for me.

Look, I’m biased, but it feels like a self-own for a “school picture day” flyer to use an AI-generated approximation of a school picture instead of an actual photograph?

The day after proudly announcing that Copilot Premium had been activated for all faculty and staff (no action required from us!), a university office has developed a form letter for how to disable it. Or at least my colleague and I got the same answer verbatim.

Got an AI-generated flyer from kiddo’s Kentucky school today that put a Texas flag in the artwork, and I have questions.

So much going on here. I think Boxer’s comments are some of the most important.

Nudifier technology is educational technology, and we can’t talk about generative AI in schools without talking about this.

Amazing reporting.

Stories like this worry me for the future of scholarship.

As I’ve said before, there’s something Ellulian in all of this AI/AI detection arms race. The creation of a different climate strikes me as more important than the utility (or problems) of any given tool.

Missed the original story, but wow.