Spencer Greenhalgh likes RSS and thinks you're great for using it

🔗 linkblog: AI detectors are creating a new era of distrust

Sun, 09 Aug 2026 10:13:03 -0400

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.

Fri, 07 Aug 2026 11:15:18 -0400

Today’s as good a day as any for an “accept with minor revisions.”

📚 bookblog: Pirate Enlightenment, or the Real Libertalia (❤️❤️❤️❤️🖤)

Sun, 02 Aug 2026 16:28:42 -0400

I thought for sure that I would be more critical of this the second time through, but I found (again?) that the last third of this is the interesting part, even if the leading up to it feels confusing.

It’s very interesting to read this after having reread Dawn of Everything and after reading There Never Was a West for the first time. I can better see the connections between those three pieces—and how they might inform my future writing.

🔗 linkblog: How a Yale AI-cheating dispute became a 13-count federal lawsuit

Fri, 31 Jul 2026 10:08:39 -0400

This story is full of twists and turns, and what I think is most important is how generative AI creates a new environment.

I think students using generative AI to cheat is super dumb. I also think that using AI detectors is super problematic. Independent of what did or did not happen here, I’m struck by the way that the mere existence of these kinds of technologies creates a multilayered climate of suspicion and distrust.

This student’s case sure seems fishy, but his distrust of AI detection tools isn’t wrong, but his university’s concerns aren’t wrong either. We can talk all we want about “right” and “wrong” uses of this tool, but the whole context had changed on a fundamental level, and that is creating problems for everyone.

A merely utilitarian perspective on technologies and ed tech doesn’t account for everything here. We need something more holistic, like Jacques Ellul’s technique to make sense of how generative AI has changed things.

🔗 linkblog: Pluralistic: The stupidest imaginable excuses for surveillance pricing (30 Jul 2026) – Pluralistic: Daily links from Cory Doctorow

Thu, 30 Jul 2026 11:39:06 -0400

I just got rotated off my grad program’s data science class, where I spent a week on the importance of theory, etc. If I were still teaching it, I’d add in this gem of a paragraph on surveillance pricing:

It’s “theory-free.” You don’t have to discover why a group is willing to pay more under a given set of circumstances, you merely have to observe and weaponize this fact. You don’t have to know why one group of purchasers consistently accept higher prices between 6AM and 8AM – you can just automatically jack up prices on them without knowing or caring that you’re gouging parents of young children who are re-ordering essential supplies while trying to get their kids to school in the morning.

📚 bookblog: The Ultimate Hidden Truth of the World... (❤️❤️❤️❤️❤️)

Sat, 25 Jul 2026 20:29:15 -0400

Rereading The Dawn of Everything finally pushed me to use an on-hand Libro.fm credit to get this in audiobook so I could give it a try. I find Graeber to be a bit uneven (I appreciate most of what he writes, but I don’t find all of his arguments totally convincing), and so I waffled on whether to give this full marks.

Ultimately, I went with the more generous rating, because even if some of his ideas didn’t land with me, it’s useful to have so much of his thinking captured in one place. In fact, I’m pretty sure I’ll need to get a print copy of this for the same research project that Dawn of Everything had me thinking about.

Of course, a lot of the essays here were good in their own right. There Never Was a West is helpful for fleshing out his idea of the indigenous critique, I find that Bullshit Jobs is stronger as an essay than it was a whole book, and the final essay on fun and play was very helpful for teasing out some of the anarchist-flavored similarities I find between his writing and that of Jacques Ellul.

new insight five years on

Fri, 24 Jul 2026 16:10:25 -0400

I’ve mentioned a few times recently across my different subblogs that importing my Twitter archive into the Day One journaling app has been interesting for my ability to go back into my past in ways that I wasn’t able to before. Yesterday, going through Day One’s “On This Day” feature turned up a tweet from five years earlier that stood out to me:

Also, this started as a silly idea while biking to campus one morning, but I increasingly want to write something on Mormonism, technology, and authority that a) discusses interplay between “tools” and “systems” (both as tech) and b) uses Hiram Page and seer stones as an example.

It’s not necessarily that I’d forgotten about the tweet—I can’t say for sure whether I specifically remember composing that post, but I do remember the feeling of having that idea. For several months, I had some rough ideas scribbled out on an office whiteboard, wondering how I might tackle that question. Rather, what stood out is the way that the post felt particularly relevant to me right now.

About a week ago, a mentioned that I had a rough idea for a book, and that rough idea is very much in the vein of my “silly idea” from five years ago. So, this is definitely an idea that has been percolating in my head for quite some time, but something important has changed in the past year or two. Rather than scribbled ideas on a whiteboard, I think I have a sense of the theoretical framework that I’d bring to bear on this question.

Perhaps unsurprisingly, that framework is an Ellulian one! I know that I’ve been a bit of a broken record about Ellul on this blog for the past year and a half, but one of the reasons for that is because his writing gives me language for things that I’ve been trying to puzzle out but didn’t have the words for. Coming across the journaled tweet is a nice way of seeing how things have come together over the past five years, and I’m excited to see where things go over the next five.

📚 bookblog: The Dawn of Everything: A New History of Humanity (❤️❤️❤️❤️❤️)

Fri, 24 Jul 2026 16:00:59 -0400

I wasn’t expecting to count this as a “work read” when I decided to reread it a couple of months ago, but wow, did it ever help make sense of some ideas I’m kicking around for a potential book project.

Like my last read, I have to admit that I’m ill-suited to evaluate the historical or anthropological claims. I really appreciate the framework it brings to its inquiry, though, not least for the way it echoes similar ideas in Jacques Ellul’s work. This gets me thinking about anarchist commonalities between these authors, which is something I want to explore more.

🔗 linkblog: The dizzying experience of 'vibe coding'

Sun, 19 Jul 2026 13:55:56 -0400

I always pay attention to Josh’s perspective on things, and this post really stood out to me.

Sun, 19 Jul 2026 11:30:24 -0400

I think I have an idea for a book, but a book doesn’t make professional sense for at least two years, BUT this means for the first time in my career, I have an idea of a true research trajectory instead of repeatedly stumbling into cool projects.

Thu, 16 Jul 2026 09:13:25 -0400

Nothing annoys me more than an academic argument whose conclusion I want to agree with but is constructed in a sloppy, superficial way.

Call for Papers: Platformed Mormonism

Mon, 13 Jul 2026 15:08:04 -0400

This CFP has been up on the journal website for a while now, but with our second “rolling deadline” of September 1st coming up soon, I thought it was time to start sharing it a little more personally. In short, I’m thrilled to joining Spencer Stewart and Rosemary Avance in editing a special issue of the Journal of the Mormon Social Science Association on Platformed Mormonism. Here’s a quick blurb on the issue, and please check out the link above for more details:

The Journal of the Mormon Social Science Association (JMSSA) is accepting submissions for a special issue exploring the intersection of Mormon traditions and digital platforms. Papers accepted for the special issue will receive an honorarium of up to USD 1,000. We welcome interdisciplinary perspectives that explore how the technological, architectural, and political structures of digital platforms are not only governing how Mormonism is perceived, represented, and encountered by the broader public but also mediating and reshaping belief, practice, and community across various Mormon traditions. We welcome both scholarly/empirical studies and shorter reflective submissions.

As I suggested above, we have a set of rolling deadlines, the next of which is September 1st, 2026, with two more on December 1st, 2026 and March 1, 2027. We encourage submitting to the earliest deadline that makes sense for the authors, but anything submitted by the March 1st deadline will be considered for the special issue. I also want to emphasize two points from the blurb above: First, that we’re accepting both longer, empirical submission and shorter, reflective submissions; and second, that papers accepted in both formats will receive honoraria. If the blurb or the longer CFP don’t answer your questions, please feel free to reach out to me specifically!

conference presentation on Jacques Ellul and Mormonism

Mon, 13 Jul 2026 09:28:31 -0400

Later this week, I’ll be presenting for the first time at the International Jacques Ellul Society. My paper is called Image and Word in Mormonism’s Foundational Media Events, and I’ll be using Ellul’s image and word framework in The Humiliation of the Word as an explanatory framework for how the founding events of Mormonism can be understood by scholars and believers as either the sure-and-certain image or the ambiguous and fleeting word. Perhaps unsurprisingly, Latter-day Saints and their leaders tend to prefer the image over the word, and I think Ellul’s writing helps understand why this is (beyond the obvious reasons)

I’m quite excited about this, given how much Ellul has affected my scholarly (and, frankly, personal) thinking in the short time that I’ve been reading him. In fact, over the past couple of weeks, as I’ve been putting final touches on my paper and slides (and doing other reading), I’ve been thinking that I’d like to use this as a starting point for a broader project on what I’m tentatively calling the “Latter-day Saint technical apparatus.” I’ve got another paper in mind for an upcoming special issue from Dialogue: A Journal of Mormon Thought, and we’ll see if I can find time in the next three months to put that together.

Anyway, for anyone outside of IJES who might be interested, I’ve created a new entry in the Research section of my website that has some information about the presentation and a link to the slides that I’ll be using.

🔗 linkblog: Suspecting AI cheating, Ivy League prof ordered an in-person final; scores fell 50%

Wed, 08 Jul 2026 19:19:03 -0400

For my entire career, I have resisted demonizing students and have sincerely believed in the need to adapt assessment to context rather than police student behavior. Yet, this kind of thing is untenable:

Eighteen students suddenly dropped the course, while nine others didn’t even attend the final exam. Of those 27 students, El País noted, “22 had scored a perfect 100 in the midterm exam.”

Tue, 07 Jul 2026 10:50:37 -0400

One perk of using Markdown/HTML-based slides is that I can work on them from my phone’s text editor when sitting a few rows back from kiddo at a summer library event.

rediscovering the library as workplace

Wed, 24 Jun 2026 09:39:01 -0400

Kiddo is at a half-day summer camp this week, which is freeing up my mornings for work. However, the camp is across town from where we live, so rather than driving all the way home for a couple of hours of work, I’ve been driving to a nearby branch of the Lexington Public Library to set up my laptop and get some work done there. It has been really nice!

When I was in grad school, I often made the library a workplace. For a while, I had a semi-regular “Spencer writes in the library” series going on my WordPress blog. I would pick a different place each week to set up shop, focus on writing, and then blog about my experience there. I might have a backup of that site that I could dig some of those posts out of (or rather, I do have a backup that I might be able to dig posts out of—I feel like I’ve done that before, but it feels opaque as I glance at it today). It has been fun to recapture that feeling of working at the library as a third place between home office and campus office.

In fact, I think I should do this sort of thing more often. Because I’m a bike commuter, getting to my campus office is a long, sometimes sweaty process. Because I try to use my professional flexibility to make room for more present and intentional parenting, I tend to prioritize working from home whenever I can rather than making it to campus to work. During May 2026, though, I felt like this meant rarely getting out of the house during the critical weeks where my semester had ended but kiddo’s hadn’t yet. This didn’t seem to be great for my mental health, which cancelled out a lot of “productivity gains” (ugh) that I had been hoping for.

If a 40-minute bike ride and ten minute change of clothes on either end of the workday doesn’t feel like a great investment when I’m trying to clear my professional checklist to make time for kiddo, and if rarely leaving the house so that I don’t need to invest that time doesn’t feel great for my mental health, maybe rediscovering the library as a workplace offers a compromise position. Biking to the closest branch of the Lexington Public Library would only take 10 or so minutes, and I wouldn’t feel the pressure there that I would at the office to change from bike commuting clothes to something more professional.

I’ve got another “critical week” coming up in August where kiddo will be in school but my semester won’t have started yet. I think I ought to look into the library as a workplace then, to better balance mental health and time to work. Before I even get there, I want to think about how to use the library as a workplace over the summer, while kiddo is at camp, or maybe letting her roam and read during a morning that I need to get a few things done.

🔗 linkblog: Pluralistic: The (real) dead economy theory (17 Jun 2026) – Pluralistic: Daily links from Cory Doctorow

Sat, 20 Jun 2026 07:50:21 -0400

Recently, I’ve been thinking about Jacques Ellul’s image/word dichotomy as more broadly applicable than to media—as applicable to the whole technical phenomenon that he criticizes. That is, by virtue of its seeming objectivity, the technical phenomenon enjoys trust that is not warranted when one digs into the details and sees the fuzziness and subjectivities and artifice that prop the whole thing up. Doctorow’s critique here seems firmly in that vein.

Thu, 18 Jun 2026 10:42:20 -0500

Currently, I am experience the specific academic joy of receiving a peer review invitation that feels like recognition of a specific research reputation I have been trying to cultivate for a few years.

🔗 linkblog: Proposed new US funding rules: We can cancel any grant at any time

Fri, 29 May 2026 21:44:53 -0400

🔗 linkblog: Proposed new US funding rules: We can cancel any grant at any time https://spencergreenhalgh.com/work/2026-05-29-i-dunno/

🔗 linkblog: ‘BusPatrol’ Put AI Cameras in Tens of Thousands of School Buses. Now They Want to Give Cops Access

Tue, 26 May 2026 12:36:27 -0400

🔗 linkblog: ‘BusPatrol’ Put AI Cameras in Tens of Thousands of School Buses. Now They Want to Give Cops Access https://spencergreenhalgh.com/work/2026-05-26-if-edtech/

📚 bookblog: A Documentary History of the Book of Mormon (❤️❤️❤️❤️🖤)

Tue, 26 May 2026 10:26:44 -0400

Reviewing reference works that I largely skimmed always feels silly, but it’s my website, and no one can stop me.

I’m not a historian, but this historical context was really useful for me for a paper I’m writing on Ellulian perspectives on “media events” in early Mormon history. I appreciated having easy access to documents related to events I was writing about, and as I eventually expand the paper, I expect that I will come back to this volume.

Morris feels a need to weigh in on apologetic/critic debates in a way that I feel is unnecessary—especially since I feel like he tries to have it both ways in pushing back against critical arguments while also saying that “the historian can’t know for sure,” etc. That’s annoying, but the volume is also so helpful for my purposes (and the commentary in mostly good faith) that I can’t penalize it too much for that.

🔗 linkblog: Researchers Wanted Preschool Teachers to Wear Cameras to Train AI

Thu, 21 May 2026 11:17:06 -0400

🔗 linkblog: Researchers Wanted Preschool Teachers to Wear Cameras to Train AI https://spencergreenhalgh.com/work/2026-05-21-very-glad/

🔗 linkblog: AI-generated research papers are overwhelming peer review

Fri, 15 May 2026 07:58:40 -0400

Here’s a gift link. Jacques Ellul argued that you can’t separate the good aspects of technique from the bad. In that context, this paragraph stands out:

Optimists about generative AI have high hopes for its ability to produce future scientific breakthroughs — accelerating discovery, eliminating most types of cancer — but the technology is currently undermining one of the pillars of scientific research, inundating editors and reviewers with an endless stream of papers. Paradoxically, the better the technology gets at producing competent papers, the worse the crisis becomes.

Even if we don’t take the super broad view that Ellul argues for, though, I think there’s still something to be said about the scale of generative AI being an issue, even if it can be used for good and bad purposes. Even if we were generous and assumed that a majority of scientific uses of generative AI were positive, two things are worth noting: First, the scale at which the minority of negative uses are carried out would still be a hell of an effect. Second, we would still struggle to respond to the positive uses at scale.

Wed, 13 May 2026 15:52:12 -0400

Marketing yourself as “the tech in edtech” is not at all a flex like Infinite Campus seems to think it is. Like, all edtech is tech—that’s the point. I don’t know that I’d agree with “the ed in edtech” either, but I’d at least get what they’re going for. This just boggles me.

Sat, 09 May 2026 11:35:02 -0400

My decision a few months ago to import my Twitter archive into my journaling app has been a mixed bag. I’ve been deleting a lot of orphaned references and truncated posts, but also, now I have a record of the time I picked a fight with an LMS CEO about privacy, and that’s fun.

🔗 linkblog: Canvas is open source, but its cloud services ransomware attack really hurts

Sat, 09 May 2026 10:00:42 -0400

🔗 linkblog: Canvas is open source, but its cloud services ransomware attack really hurts https://spencergreenhalgh.com/work/2026-05-09-bens-perspective/

Fri, 08 May 2026 16:56:02 -0400

It’s amazing how much I struggle to write a single sentence during time I’ve set aside for writing, but when I’m in the middle of grading, I can “pause for a second” to jot down an idea and wind up with three draft paragraphs.

Fri, 08 May 2026 14:21:54 -0400

I love a genuinely good experience as a peer reviewer. The current manuscript I’m reviewing is one that I’m very well suited to review (which does not happen enough), and the authors have clearly improved the paper since the last round. It’s nice when things work like they’re supposed to.

🔗 linkblog: 'The Biggest Student Data Privacy Disaster in History': Canvas Hack Shows the Danger of Centralized EdTech

Fri, 08 May 2026 12:22:53 -0400

🔗 linkblog: 'The Biggest Student Data Privacy Disaster in History': Canvas Hack Shows the Danger of Centralized EdTech https://spencergreenhalgh.com/work/2026-05-08-some-important/

🔗 linkblog: The Canvas Hack Is a New Kind of Ransomware Debacle

Fri, 08 May 2026 06:49:55 -0400

Definitely worth bookmarking this for semesters to come. I like Canvas as far as LMSs go, but the sheer scale of dependence here has me thinking about taking other approaches. This first paragraph is a doozy:

Higher education has long been a target of ransomware gangs and data extortion attacks. But never before, perhaps, has a cyberattack against a single software platform so thoroughly disrupted the daily operations of thousands of schools across the United States.

🔗 linkblog: Canvas is down as ShinyHunters threatens to leak schools’ data

Thu, 07 May 2026 18:46:08 -0400

🔗 linkblog: Canvas is down as ShinyHunters threatens to leak schools’ data https://spencergreenhalgh.com/work/2026-05-07-oh-wait/

Thu, 07 May 2026 18:42:37 -0400

Amazing time of year for this to happen.

🔗 linkblog: Book publishers sue Meta over AI’s ‘word-for-word’ copying

Tue, 05 May 2026 16:04:38 -0400

This is a good example of how thorny the AI problem is, and why I strongly prefer a digital labor critique to a copyright critique. Yes, I’m mad that Meta trained their models on my work, but I don’t think the answer is to strengthen Elsevier or Cengage’s copyright claims.

🔗 linkblog: University Professors Disturbed to Find Their Lectures Chopped Up and Turned Into AI Slop

Mon, 27 Apr 2026 19:17:25 -0400

🔗 linkblog: University Professors Disturbed to Find Their Lectures Chopped Up and Turned Into AI Slop https://spencergreenhalgh.com/work/2026-04-27-well-this/

🔗 linkblog: Prestigious photo contest answers ‘what is a photo?’

Fri, 24 Apr 2026 08:26:29 -0400

I wish I knew enough about photography to really appreciate the details here, but I’m bookmarking it anyway because it feels like a contemporary, unintentional echo of observations Ellul makes in The Humiliation of the Word.

Wed, 22 Apr 2026 09:11:34 -0400

I’ve been thinking for a while that the connection between John Perry Barlow and Marc Prensky re: the mythical “digital native” is underexplored, and this morning, I finally started a folder for a new writing project.

🔗 linkblog: Beshear is ‘losing confidence’ in University of Kentucky admin following recent hirings • Kentucky Lantern

Tue, 21 Apr 2026 15:39:13 -0400

You know, I wasn’t sure about the choice of Van Tatenhove as the next law school dean, but I figured that I didn’t know that much about the guy, and if the law faculty was on board, I ought to trust that rather than my gut reaction.

Finding out that faculty were not on board and that the university seems to have made the decision from above despite faculty concerns is more worrying than any concern I previously had.

🔗 linkblog: The Deepfake Nudes Crisis in Schools Is Much Worse Than You Thought

Wed, 15 Apr 2026 07:37:00 -0400

Audrey Watters once compellingly argued that metal detectors are edtech. I think we now have a responsibility to treat AI nudifier apps as edtech, too.

🔗 linkblog: To teach in the time of ChatGPT is to know pain

Mon, 13 Apr 2026 07:43:46 -0400

Really appreciate this essay. It puts things nicely and has the kind of personal investment that makes it relatable.

Fri, 10 Apr 2026 12:03:22 -0400

I am in my late 30s. “Digital native” was first used for my age cohort in the early 2000s—and was riffing off of John Perry Barlow’s 1996 “Declaration of Independence of Cyberspace.” It is a USELESS term because its frame of reference keeps shifting to “kids these days” (and other reasons, too).

🔗 linkblog: UK hosts literacy training in AI to teach attendees of its potential

Thu, 02 Apr 2026 18:46:10 -0400

Two grumps (and, to be clear, I’m grumpy at my employer, not the student reporter):

the framing here is, as usual, “how to use” whether than “should we use”
“misinformation” is centered as the (implicitly sole) problem with generative AI, not digital labor or any of the deeper issues

🔗 linkblog: Sweden goes back to basics, swapping screens for books in the classroom

Wed, 01 Apr 2026 13:03:45 -0400

Mixed feelings about this. I think there are good reasons to be skeptical of ed tech at this level, but I also think that there is some unwarranted handwringing going on here. I have trouble untangling the two.

Jérôme Dupras, my favorite bassist-drummer-academic

Mon, 30 Mar 2026 10:35:08 -0400

Les Cowboys fringants is a Québécois folk rock band that has gotten a lot of my attention over the past five years (though I’ve enjoyed their music for even longer than that. I have a bunch of their albums, but over the past few months, I’ve taken to listening to some of their concerts that have been recorded. I do this while I’m working, so I’m usually listening to the sound rather than watch the visuals, but I do enjoy taking the occasional peek at what’s happening on stage, too.

There’s a moment in a 2023 concert where Jérôme Dupras, who is the band’s bassist and drummer, does a neat little drum solo before goofing around on stage for a few minutes, then bringing out a ladder, putting it at the end of the stage, climbing to the top, tearing off his shirt, and then diving into the audience for a bit of crowd surfing. As he’s being brought back to the stage, lead singer Karl Tremblay (who will pass away later that year from cancer) has this to say:

C’était le docteur Jérôme Dupras, mesdames, messieurs. Un homme qui fait des conférences à travers le monde, un scientifique imminent et très respecté, qui a fait ça pour vous ce soir—se dénuder et plonger dans la foule.

In English:

That was Dr. Jérôme Dupras, ladies and gentlemen. A man who attends conferences all over the world, an up-and-coming, respected scientist, who did that for you—strip down and dive into the crowd.

This cracks me up because it is 100% true. When he’s not playing bass, doing drum solos, and crowd surfing at a show for one of Québec’s most famous musical groups of all time, Dupras is a professor of ecological economics—as well as a Canada Research Chair and UNESCO Chair—at the Université du Québec en Outaouais. He joined the band as a novice bassist before starting his academic career, making his way through his studies and into academia at the same time that the band was getting bigger and bigger. He has stories to tell of writing grant applications on tour buses, and he’s certainly the reason that this is one of the few bands in the world to have a song that casually references a character going off and doing a post-doc somewhere. If you search for his name on YouTube, you get a mix of him appearing on the news to talk about his research and other clips of drum solos and crowd surfing.

I’ve achieved most of what I really ever wanted to get out of my career, and I’ve had some recent experiences that remind me that the most important thing I want to accomplish in this job is to be remembered as someone who was kind and supportive, not someone who was difficult to work with. In the moments where I do allow myself a bit of professional jealousy, though, I often think of Dr. Dupras, who has not only accomplished more in his academic career than I ever will, but has also done it while taking the occasional break to tour with a band that has made French Canadian history. What a cool guy.

what I dislike about AI isn't the tech (and why I like Ellulian 'technique')

Tue, 24 Mar 2026 12:32:47 -0400

Last Thursday, I listened to a recent episode of The Vergecast during my morning bike commute. The episode featured Paul Ford talking about his recent experience with Claude Code, and I was genuinely surprised to find some of his comments resonating with me. It helped that Ford wasn’t uncritical about AI (though certainly not as critical as I would have been), but some of it was just that I recognized some of the thrill that he was describing of using tools and resources to learn how to solve a problem. In fact, I found that thrill so contagious that a passing comment he made got me to spend some time once I got to the office converting my Twitter archive into a CSV that I could finally import it into the Day One journaling app that I use.

Of course, I did all of that work by hand, because I’m still at a point (and hope to stay at a point) where I refuse to use generative AI tools to contribute to my work. Nonetheless, listening to the interview really got me thinking about what it is about generative AI that bothers me so much—and it’s pretty clear that it’s not the technology itself. For all of their hallucinations and similar issues, image generation tools, LLMs, and the like are technically impressive! Just in terms of what they can do, they are pretty neat technological developments, and for all of my grumpiness, I’m happy to concede that.

So, what makes me angry about generative AI isn’t the technology itself, it’s all the baggage that it brings with it. It’s the extractive nature of how the models are trained. It’s the scattershot, unreflective way that the technologies are being rolled out. It’s the insistence that this is the future and that everyone needs to get on board. I wouldn’t necessarily be opposed to a carefully and consensually trained model that was applied in specific circumstances without being imposed on anyone—that could be a really neat application of these underlying technologies. That’s not the generative AI that we’re being presented with, though, and my abstract appreciation for “neat tech does neat things” doesn’t survive contact with the status quo.

Thinking about all of this reminds me why I appreciate the work of Jacques Ellul as a theoretical framework for thinking about technology. (Today, coincidentally, marks the first anniversary of my having finished his The Technological Society). While Ellul is often (and not incorrectly) described as a critic of technology, it is more accurate to describe him as a critic of technique, a term with different connotations. Some of these differences are due to linguistic and cultural differences between the Anglo-Saxon world and Continental Europe, but Ellul also had an idiosyncratic way of using the term that distinguishes him from other French (and Continental) writers.

In short, technique is more than any given tool or technology. First, Ellul insists that the entire technological system—the (translated) title of another of his books—be considered together instead of as individual pieces. Second, Ellul emphasizes that societal trends and attitudes—chiefly, an unyielding insistence on efficiency and efficacy—are part of that technological system. In fact, it’s easy to miss among his fierce criticisms that Ellul is not necessarily opposed to individual tools and technologies. His critique is most specifically oriented at that societal trend, at the insistence that one must adopt all new technologies because they are more efficient. Here’s a taste of that in the posthumously published Théologie et technique, where Ellul argues for an ethic of non-power, in which people deliberately choose not to do everything that they’re capable of and are more deliberative about what makes sense (again, not just what is possible—it’s a very Ian-Malcolm-in-Jurassic-Park ethic):

Le système technicien ne peut strictement pas supporter une attitude de vie de non-puissance, ce serait sa ruine : il suffit de penser que l’on ne choisirait plus, pour consommer, ce qui est le plus rapide, le plus efficace, le plus perfectionné. Je ne dis pas que l’on choisirait systématiquement le moins rapide, le moins efficace, etc., mais simplement on cesse d’être intéressé par ces qualités.

Here it is in my rough translation

The technological system strictly cannot tolerate an attitude of non-power toward life, for it would be its ruin: It suffices to think that one would no longer choose, when consuming, what is the fastest, the most efficient, the most perfect. I do not say that one would systematically choose the slowest, the least efficient, etc., but simply that one would cease to be interested by these qualities.

Ellul’s technique is helpful for me in understanding my frustration with generative AI. It’s not the tool itself, it’s the broader system that surrounds it. If generative AI could be separated from the societal attitudes of “what can be done must be done” and “what seems more efficient is necessarily better,” I think I could get on board with the tool in a way that I am currently not. However, 30 years after Ellul’s death, I feel like generative AI more than anything else shows the utility of this theoretical framing of this, and other, technologies.

Fri, 20 Mar 2026 13:55:05 -0400

Once again, my spring break has consisted of: trying to catch up on work, telling myself that it’s spring break and therefore okay to not catch up on as much work as I expected, and feeling guilty about not catching up on as much as I expected

Mon, 16 Mar 2026 14:11:06 -0400

Oh, it’s not just in RSS!! Page is still up!

Mon, 16 Mar 2026 14:01:29 -0400

Love it when this kind of thing shows up in my RSS reader so I can use it as an example in my content management systems class.

Thu, 12 Mar 2026 09:21:15 -0400

Just checked up on the results of a web scrape that worked so, so much better than expected only to find that I hadn’t updated the base URL correctly, so all the results were wrong.

🔗 linkblog: What’s the Point of School When AI Can Do Your Homework?

Wed, 25 Feb 2026 11:55:14 -0500

🔗 linkblog: What’s the Point of School When AI Can Do Your Homework? https://spencergreenhalgh.com/work/2026-02-25-the-headline/

Wed, 18 Feb 2026 13:14:55 -0500

There are so many layers of unearned, inaccurate, and problematic chatbot anthropomorphism here that I honestly don’t know where to start.

Wed, 18 Feb 2026 10:25:36 -0500

Grading anxiety (“am I not teaching this right? is my rubric unclear?”) has been a persistent part of my academic career for so long that I’ve had to deliberately embrace that Simpsons Skinner meme and tell myself that sometimes “no, it’s the [students] who are wrong.”

Wed, 18 Feb 2026 07:28:30 -0500

TFW a special issue deadline gets extended by a month after you dropped everything for a week to meet the original deadline. Alas.

🔗 linkblog: 'Students Are Being Treated Like Guinea Pigs:' Inside an AI-Powered Private School

Tue, 17 Feb 2026 11:30:06 -0500

🔗 linkblog: 'Students Are Being Treated Like Guinea Pigs:' Inside an AI-Powered Private School https://spencergreenhalgh.com/work/2026-02-17-so-many/

Thu, 12 Feb 2026 09:45:01 -0500

In a Zoom meeting with the dean, with my X-Wing and Enterprise-D posters clearly visible in the background. Living my most professional life.

Ellul strikes again

Tue, 03 Feb 2026 15:54:29 -0500

I began my sudden but immediately sustained interest in Jacques Ellul about a year ago now, and I’ve found his work to be terribly influential on my personal thinking and my professional work. I’m currently working on a manuscript that makes the argument that Ellulian thought is useful for drawing our attention in certain ways when considering artificial intelligence in education. I see theory as serving an analytical and rhetorical purpose for the way that it makes suggestions that a certain phenomenon works in certain ways and invites us to consider whether or how that is true.

One of the consequences of all of this is that I tend to see things through an Ellulian lens right now, asking myself what in such-and-such a news story corresponds with arguments he made about technology (or, more accurately, technique). If you scroll through the “Jacques Ellul” tag on my website, you’ll see that that’s happening a lot recently.

This post is sparked by a microblog post that someone wrote over the weekend about the Moltbook “social media platform for AI agents” that’s been in the news a lot lately. Here’s the text that stood out to me:

I’m not saying it’s good or bad… Value judgements don’t even matter right now.

This is a remarkably Ellulian turn of phrase—or at least one that he attributed to others and then critiqued. Here he is in a 1980(!) book chapter, “The Ethics of Nonpower”:

Technique itself has become a value. Technical progress appears to the average Western person as the guarantee of the future good and happiness, and technology assures him of the necessity of the kind of behavior favorable to this progress. Technique carries our hopes (thanks to technical progress, cancer will be conquered). Here it gives life a meaning. And the usual attitude, whenever there appear to be drawbacks in the use of technique, consists in declaring that it is not technique that is to be blamed, but rather man, who does not know how to use it. This means, by implication, that it is man who produces evil and that technique therefore stands for good.

He writes more on this in other sources as well, but a lot of it comes down to this: That technological development is seen as a self-evident good that somehow transcends considerations of good and evil. I don’t want to read too much into a single microblog post, especially since I respect its author even when I disagree with his takes, but it’s remarkable to me to see the attitude that Ellul was criticizing reproduced so faithfully 45 years after Ellul wrote that paragraph.

Sat, 31 Jan 2026 09:47:59 -0500

Can’t even start a lazy Saturday archive binge of a decades-old webcomic without coming back to Jacques Ellul, concerns about quantification and metrics, etc.

Wed, 28 Jan 2026 15:10:50 -0500

Look, I genuinely appreciate my university employer sending this out in their daily email, but it would be cool if they also acknowledged how much data they and their various platform partners are collecting on faculty and students. Because it’s a lot.

Tue, 27 Jan 2026 18:41:54 -0500

A faculty candidate visit for tomorrow just got moved 100% virtual, and while this sucks for all of us in so many ways, it also means I don’t have to clean out my car for fear of being embarrassed by the mess when giving the candidate a ride.

Fri, 23 Jan 2026 17:41:28 -0500

Kiddo was asking me questions about the manuscript I was writing, that got us talking about science fairs, and then she basically reasoned her way into the idea of a meta-analysis.

🔗 linkblog: Google’s work in schools aims to create a ‘pipeline of future users,’ internal documents say

Fri, 23 Jan 2026 15:36:49 -0500

Wish I’d had this to cite in some recent publications. What a great(?) example of saying the quiet part out loud:

One internal November 2020 presentation slide said acclimating children to Google’s ecosystem in school would hopefully lead them to use its products as adults: “You get that loyalty early, and potentially for life.” Another undated slide deck suggested imagining a world where “Parents ask their children ‘Why aren’t you watching more YouTube?’” and “School Administrators shift budgets from Textbooks to YouTube subscriptions.”

🔗 linkblog: New AI-Generated Content Derived from Your Work Posted on Academia.Edu

Wed, 21 Jan 2026 13:15:22 -0500

I guess I should be reading this for the jokes, but I hadn’t realized Academia.edu had done this, and I’m so angry at the inspiration for the jokes that I haven’t made it any further.

Tue, 13 Jan 2026 13:55:20 -0500

Months ago, I submitted a paper to a CFP outside my area—and immediately wrote my co-author a list of what I thought probably still needed fixing. Reviews came back today with both reviewers recommending “ready for publication.” Academic writing is unpredictable, but I’ll take it in this case.

🔗 linkblog: Pluralistic: Writing vs AI (07 Jan 2026) – Pluralistic: Daily links from Cory Doctorow

Wed, 07 Jan 2026 11:47:07 -0500

I have largely abstained from the “AI misses the point of writing” discourse, but Cory knocks it out of the park here.

Wed, 31 Dec 2025 11:35:14 -0500

Prepping one of my Spring courses, and I’m apparently so tired that I’ve just decided to appeal directly to my students’ sense of decency when it comes to respecting course policies on generative AI.

digital labor and generative AI: what Stack Overflow CEO Prashanth Chandrasekhar gets wrong

Wed, 17 Dec 2025 08:55:05 -0500

This morning, while getting ready for the day, I spent some time catching up on podcasts, including Nilay Patel’s interview of Stack Overflow CEO Prashanth Chandrasekhar on a recent episode of Decoder (a podcast I’ve spent a lot more time listening to since it went ad free for subscribers). I ditched the Stack Exchange network a year and a half ago over digital labor concerns—I was literally being prevented from deleting my own content from the site, which is bonkers—and I’m honestly not sure why I bookmarked the interview for listening a few days ago. I think it was more than a hate listen, though: For all of my own feelings about generative AI, I make an effort to be open minded, and I was interested in the headline for the interview: “Stack Overflow users don’t trust AI. They’re using it anyway.”

One exchange between Patel and Chandrasekhar really stood out to me, though, and not necessarily in a good way. I was pleased that Patel pushed Chandrasekhar on the question of digital labor, but the CEO’s response to Patel’s question really rubbed me the wrong way. Here’s what Patel had to say:

If I am somebody in your 1 percent who spends a lot of time on Stack Overflow helping other people. The reason I answer questions for free on your platform, which you monetize in lots of ways, is because I can directly see that my effort helps other people grow and that I’m helping other people solve problems. That is one very self-contained dynamic. The last time you were on the show, our entire conversation was about that dynamic and how you got people to participate in that dynamic and the value of it.

Then suddenly, there’s a very clear economic benefit to the company that owns the database because it’s selling my effort to OpenAI, which is happening across the board. It’s going to do these data licensing deals with all these AI providers, they’re going to train on the answers that I have painstakingly entered into this database to help other people, and now the next generation of software engineers is going to get auto-complete that’s based on my work and I’ve gotten nothing. I’ve heard that from lots and lots of people. I’ve heard that in our own community, and I think I have felt that as various media companies have made these deals.

How do you respond to that? Because it feels like you were providing a database that you had to monetize in some ways, but the interaction people had was the value, and now there’s another kind of economic value that is maybe overshadowing, recasting, or re-characterizing the interaction that people have.

Here’s, then, how Chandrasekhar responds, which is an answer smooth enough that it almost carries weight if you don’t think any more about it:

There are a couple of points there. One is about this company’s original DNA and why people came together to do this thing. When I joined the company, I asked a question like, “What’s people’s incentive to spend time doing this?” I asked the founders, specifically [co-founder] Joel Spolsky, about this. His point was that the software development community is very altruistic. People just want to help each other out because people understand how frustrating… I used to write code many years ago. I recently picked it back up with some of the code-generation tools, which is interesting to compare and contrast. I just remember how frustrating it was if you got stuck on something. Stack was a huge boon when it was created to unlock this. It was truly out of that. That was the reason.

Even before ChatGPT, we also asked the question, “Should we incentivize users by paying them? Should we give them a monetary benefit?” That wasn’t a high ask by a user base. We went and researched people. People were not in for the money. Plus, it complicates things because how do you judge the payment for a particular JavaScript question relative to a particular Python question? It goes down a rabbit hole, which is untenable. So that’s one. What was the original reason people got together; it was about the mission.

Here’s the thing, though: Not wanting to get paid yourself is not blanket permission for other people to profit off of your work. I feel this most pressingly as an academic (which is why I’m writing this on my workblog instead of one of my other subblogs). I don’t necessarily feel the need to get royalties every time someone accesses, reads, or cites one of my scholarly publications. I need a paycheck, sure, but I get a paycheck from the University of Kentucky for being a researcher, and I think the model of being paid by public tax dollars (and a crapton of student tuition, but that’s another problem for another post) to generate knowledge and release it freely into the public domain is a good one! (I also acknowledge that this answer might be more complicated for an independent researcher or a scholar in a more tenuous professional position).

Here’s the thing, though: If the ideal of my salary coming from public tax dollars to do this work doesn’t quite hold up, the idea that I release the knowledge I generate into the public domain is straight up laughable. I would be happy—ecstatic even—to release my publications into the world and to never get paid for them, but that’s not how it works. Instead, academic publishing companies claim my labor and the labor of peer reviewers, require me to sign over copyright (or at least some kind of exclusivity claim), and then proceed to charge my own university for access to my work—often at a high profit margin. This strikes me as unacceptable—and that’s before we get to the question of generative AI companies pirating my work from those companies to profit from my labor in their own ways.

So, contrary to what Chandrasekhar argues, there’s a difference between wanting to get paid for one’s work and objecting to others’ profiting off of that work. That one does not need to get paid for their labor does not mean that one consents to others’ profiting off of that same labor. If I were better read on the gift economy, I’m sure I could draw some insightful parallels there, but let me instead use an analogy that I think gets at what I’m trying to say. Let’s imagine that after some kind of natural disaster, Aïcha is in a position of relative privilege and is giving away water bottles and canned goods for free, no questions asked, at a table she’s set up in a public park. Bertrand comes up to Aïcha and claims an armful of water and food, thanks Aïcha, and walks away. Bertrand sets up his own table at a different public park and proceeds to sell the goods he’s received from Aïcha to those who are walking by.

When Aïcha discovers this and gets upset at Bertrand, the problem is not necessarily that Aïcha did not get paid for the water and food that she distributed. She was happy to give the goods away, and she might even feel that it isn’t right for her to accept money from Bertrand (or anyone else) for what she is doing. The problem is that Bertrand has violated the social contract that Aïcha believed that everyone was operating under—he sought to profit off of others’ generosity. When explained in these terms, I don’t think Chandrasekhar’s answer actually addresses Patel’s question (though perhaps Patel could have asked a slightly different question, too).

There’s more I could write about this interview and what parts of which answers I objected to, but this point strikes me as parallel to the “copyright vs. digital labor” issue that I’ve been trying to tease out in recent points, and I thought it was really worth making. My objection (and, presumably, the objection of many Stack Overflow users) to my labor being used to train generative AI is not that I am not being paid for it—it is that others are finding ways to get paid off of it. Chandrasekhar’s defense of the second by way of pointing out the first strikes me as an incomplete answer that does not actually address the problem.

Thu, 11 Dec 2025 06:32:49 -0500

Woke up this morning from a dream about curriculum redesign. How’s your end of semester going?

🔗 linkblog: Details of Mark Stoops’ buyout agreement with Kentucky revealed

Tue, 09 Dec 2025 12:37:54 -0500

If these numbers are right, the first (“small”) installment alone could potentially be higher than the total of all my paychecks between being hired at UK and my eventual retirement—and this is just a severance pay package.

Tue, 09 Dec 2025 11:21:42 -0500

Today, on the last day of class, a student shared some of their ongoing efforts to mess with local Flock cameras and develop makeup patterns to defeat facial recognition. It was like realizing I’d had a Cory Doctorow character in class all semester, and it brought me joy and hope.

Mon, 01 Dec 2025 10:17:11 -0500

My university just parted ways with our football coach for undesirable results despite the investment of millions of dollars, so I wonder if they’ll be willing to do the same for all the AI initiatives they’ve recently announced.

🔗 linkblog: UK among first universities to collaborate with Microsoft on AI

Tue, 25 Nov 2025 19:33:02 -0500

🔗 linkblog: UK among first universities to collaborate with Microsoft on AI https://spencergreenhalgh.com/work/2025-11-25-this-just/

🔗 linkblog: NKU to cut down remote work, offer buyouts to tenured faculty • Kentucky Lantern

Mon, 24 Nov 2025 06:15:00 -0500

🔗 linkblog: NKU to cut down remote work, offer buyouts to tenured faculty • Kentucky Lantern https://spencergreenhalgh.com/work/2025-11-24-hope-this/

🔗 linkblog: OpenAI just released ChatGPT for teachers.

Thu, 20 Nov 2025 19:29:12 -0500

🔗 linkblog: OpenAI just released ChatGPT for teachers. https://spencergreenhalgh.com/work/2025-11-20-le-sigh./

🔗 linkblog: He got sued for sharing public YouTube videos; nightmare ended in settlement

Wed, 19 Nov 2025 18:20:29 -0500

🔗 linkblog: He got sued for sharing public YouTube videos; nightmare ended in settlement https://spencergreenhalgh.com/work/2025-11-19-very-happy/

Wed, 19 Nov 2025 16:59:10 -0500

Another suspected use of generative AI, another decision that I still won’t subject my students to surveillance and policing, another step towards complete grading nihilism.

🔗 linkblog: Anti-DEI compliance continues across campus; written policy is hard to come by

Sat, 15 Nov 2025 07:38:17 -0500

🔗 linkblog: Anti-DEI compliance continues across campus; written policy is hard to come by https://spencergreenhalgh.com/work/2025-11-15-appreciate-the/

🔗 linkblog: UK launches CATS AI to advance artificial intelligence across campus

Thu, 13 Nov 2025 20:06:40 -0500

🔗 linkblog: UK launches CATS AI to advance artificial intelligence across campus https://spencergreenhalgh.com/work/2025-11-13-i-am/

Sat, 08 Nov 2025 21:46:11 -0500

I know I’m currently inclined to see everything through an Ellulian lens, but I think one lesson of the first episode of Pluribus is that not all science needs to be done, and especially not just because it can be done.

🔗 linkblog: The problems with AI in schools

Thu, 06 Nov 2025 12:40:00 -0500

🔗 linkblog: The problems with AI in schools https://spencergreenhalgh.com/work/2025-11-06-really-enjoyed/

🔗 linkblog: Tech companies don’t care that students use their AI agents to cheat

Tue, 04 Nov 2025 09:43:35 -0500

🔗 linkblog: Tech companies don’t care that students use their AI agents to cheat https://spencergreenhalgh.com/work/2025-11-04-adding-some/

📚 bookblog: Mediated Mormons: Shifting Religious Identities in the Digital Age (❤️❤️❤️❤️🖤)

Thu, 30 Oct 2025 17:23:46 -0500

I have been meaning to read this book for months and was happy to have an uninterrupted couple of hours on a flight so that I could finally get to it.

It covers familiar ground but without feeling repetitive. I expect to cite it frequently in the future—in fact, I ought to work it into some slides I’m presenting tomorrow!

🔗 linkblog: UK must be ‘partner-of-choice’ in using AI to advance Kentucky

Tue, 21 Oct 2025 16:53:38 -0400

Honestly trying to figure out whether the reason I see Ellul everywhere is because I’m excited about a new scholar I’ve discovered or because his ideas are so well suited for the current moment. “We can be a leader or we can be left behind” captures the opt-in determinism of Ellul’s technique so dang well.

Of course, how the heck am I going to keep expressing concern about AI (through an Ellulian lens or otherwise) if the university has already decided that we’re all getting on board?

🔗 linkblog: The AWS Outage Was a Nightmare for College Students

Tue, 21 Oct 2025 12:29:39 -0400

Universities are too platform dependent, and even those platforms are too dependent on the next layer of the stack. What a mess.

Fri, 17 Oct 2025 17:33:27 -0400

It’s not that I think pumpkin spice tea pairs well with Friday night sushi takeout, it’s that I can’t remember the last week I lost this much sleep over work anxiety, so I need dinner tea to get through the rest of the day.

Fri, 17 Oct 2025 09:47:13 -0400

Thinking this morning about all of the small, human-oriented things I’ve done this semester to try to be good at my job, and how insulting it will be to have them eventually reduced to a number that can be reported to demonstrate “productivity.”

🔗 linkblog: UK quietly dismantles gender-inclusive housing and ties to identity-based groups

Thu, 16 Oct 2025 21:05:39 -0400

🔗 linkblog: UK quietly dismantles gender-inclusive housing and ties to identity-based groups https://spencergreenhalgh.com/work/2025-10-16-good-reporting/

Wed, 15 Oct 2025 17:35:55 -0400

Remembering the time that an article in the Daily Caller took a colleague’s research out of context to make it look bad, but because it tripped a Google Alert, it still got included in some “our research in the media!” newsletter.

Mon, 13 Oct 2025 10:54:35 -0400

Wild to teach data visualization so shortly after reading Jacques Ellul rail against it in his “Humiliation of the Word.” As is often the case, Ellul’s complaints feel overstated, but as I talk through specific things with students, I find myself channeling his concerns.

Fri, 10 Oct 2025 09:56:03 -0400

“AI is as good as any rubric” (an actual statement I overheard at a campus-wide meeting), if true, is more of an indictment of current assessment practice than praise of AI quality.

🔗 linkblog: He Wrote a Book About Antifa. Death Threats Are Driving Him Out of the US

Thu, 09 Oct 2025 12:50:07 -0400

🔗 linkblog: He Wrote a Book About Antifa. Death Threats Are Driving Him Out of the US https://spencergreenhalgh.com/work/2025-10-09-well-this/

🔗 linkblog: What the Arrival of A.I. Video Generators Like Sora Means for Us

Thu, 09 Oct 2025 07:01:10 -0400

Strong Ellul vibes in this passage:

The tech could represent the end of visual fact — the idea that video could serve as an objective record of reality — as we know it. Society as a whole will have to treat videos with as much skepticism as people already do words.

Unclear, though, whether Ellul would be cool with increased skepticism of the image or angry at the technology causing it.

Here’s a gift link.

Jacques Ellul contre l'appli Sora

Wed, 08 Oct 2025 10:52:02 -0400

Un peu par hasard, j’ai fini récemment ma lecture de deux livres différents par Jacques Ellul : Théologie et technique ainsi que Humiliation of the Word (la traduction anglaise de La parole humiliée, car je vais devoir en écrire en anglais, et j’avoue en plus que mon français n’est pas toujours à la hauteur d’Ellul « en V.O. »). Ça fait plusieurs jours que j’ai envie d’écrire quelque chose sur la relation image-parole qu’il établit dans les pages de La parole humilié, et je compte toujours écrire ce post-là, mais en terminant Théologie et technique, j’ai été frappé par un passage qui ressemble beaucoup ce dont j’avais envie d’écrire dans l’autre livre.

Voici donc un petit passage à la page 330 de Théologie et technique, où Ellul écrit d’un…

…irréel qui se fait prendre pour la seule réalité, d’une image qui n’est jamais qu’image, mais se fait prendre pour la substance même, d’une représentation qui est finalement représentation de rien…

Il était impossible pour moi de lire ce passage sans penser à la nouvelle application Sora d’OpenAI, qui permet aux utilisateurs de générer facilement des vidéos assez convaincants de n’importe quel (ou presque) sujet ou personnage. Il me semble que c’est justement un irréel qui veut se faire prendre pour la seule réalité !

En plus, comme j’écrirai davantage quand je trouverai le temps d’écrire un post plus long, si j’ai longtemps hésité à adopter une perspective vraiment critique sur les médias de l’image (comme celle d’Ellul ou de McLuhan ou de Postman), c’est parce que je ne voyais pas une grande rupture entre l’image et la réalité qu’elle prétendait représenter. Je commence à voir des choses différemment (grâce à Donald Trump, d’ailleurs), mais même si je restais sceptique, je devrais quand-même reconnaître qu’une appli comme Sora, où cette rupture est bien évidente.

J’ai dit à plusieurs collègues récemment que l’IA générative, c’est la technologie qui brise mes perspectives sur la technologie et les médias et qui m’exige de nouvelles perspectives. En voici un grand exemple, et je suis content de pouvoir citer Ellul en bâtissant ces nouvelles perspectives.

🔗 linkblog: Libraries Can’t Get Their Loaned Books Back Because of Trump’s Tariffs

Mon, 06 Oct 2025 14:16:33 -0400

🔗 linkblog: Libraries Can’t Get Their Loaned Books Back Because of Trump’s Tariffs https://spencergreenhalgh.com/work/2025-10-06-this-never/

Fri, 03 Oct 2025 10:27:03 -0400

I have gotten a remarkable amount of work done on my phone while waiting for my delayed eye doctor’s appointment, but it is now time to read Rogue Squadron on Libby and not feel guilty about it.

🔗 linkblog: New tech bolsters UK's law enforcement presence, but one priority requires buy-in from campus residents

Thu, 02 Oct 2025 18:23:51 -0400

🔗 linkblog: New tech bolsters UK's law enforcement presence, but one priority requires buy-in from campus residents https://spencergreenhalgh.com/work/2025-10-02-5000-cameras/

🔗 linkblog: Research, curriculum and grading: new data sheds light on how professors are using AI

Thu, 02 Oct 2025 07:00:36 -0400

Surprised that more isn’t made of the fact that Anthropic was surveilling users’ conversations for its research. Are professors and students thinking about the company’s ability to read everything they type?

Wed, 01 Oct 2025 11:11:14 -0400

Saw an AI “summarize this discussion” button in Canvas today and it was a real downer. It seems to be only instructor-facing, which is better(?) but still makes me angry.

where I'm cited on Wikipedia

Wed, 24 Sep 2025 11:26:36 -0400

Last week, I read a post from Andrew Heiss on Bluesky that inspired me to take a look at whether/where I was cited on any Wikipedia articles. I knew my research had been referenced on one particular page, but I’d never done a thorough search for this and decided to give it a whirl.

While I can’t claim anything as cool as the page on Hosni Mubarak (where Andrew’s research is cited), my research is referenced on three different Wikipedia articles, which feels pretty cool, actually. It seems like my newer Mormon Studies work is what is getting traction on Wikipedia, as opposed to my historical (and continuing) focus on educational technology research. This doesn’t totally surprise me; I’ve observed for a couple of years (or longer) that while my edtech research gets a whole lot more scholarly attention, my Mormon Studies work tends to get more media and popular interest. I attribute this to doing niche work on subjects where a smaller number of people show a greater amount of interest in what I’m studying.

Anyway, here are the three articles on Wikipedia where my work gets cited in some way:

DezNat

This article on the online “DezNat” movement cites the article I published with Amy Chapman on the movement (to my knowledge, the most extensive scholarly treatment on DezNat) as well as a blog post I wrote where I reproduced a passage of a Salt Lake Tribune article I’d been quoted in. I have some quibbles with how my work is cited here: First, the article is cited in two different ways, not counting a citation of a Times & Season blog post that reproduces the article’s abstract. Second, it uses our research as support for describing DezNat as a “far-right” movement even though we emphasize the (intentional) ambiguity of that status. Note that this isn’t a defense of DezNat (either in the article or in this post), just an acknowledgment that at least from the theoretical perspective we used, it’s the ambiguity of the movement’s far-right status that is the most interesting. For all those quibbles, though, I’m genuinely flattered to be referenced in this article.

Mormonism and violence

I basically only appear in this article because it also references DezNat, so not much more to add here.

Church of Jesus Christ in Zion

This article on an obscure Latter Day Saint denomination/group was the one that really surprised me. Not too much, I guess: The denomination gets referenced in the article in Dialogue where I identify it as the previous owner of the LDS Church’s current domain name. That article gets cited, as does an article in the Salt Lake Tribune about my research. That said, all I know about this denomination I got from other sources, so I think that Steve Shields really ought to get the credit here.

Tue, 23 Sep 2025 06:56:55 -0400

Outlook reactions continue to be THE WORST.

Tue, 16 Sep 2025 18:15:06 -0400

Can we agree that swearing at one’s email inbox counts as work so that I can say I’ve been especially productive today?

October 6th webinar on 'dark side of affinity spaces' research

Mon, 15 Sep 2025 12:46:41 -0400

A few months ago, I announced a new publication from Dan Krutka and me based on our study of a teacher’s group on far-right social media. I’m happy to share that Renee Hobbs’s Media Education Lab has taken interest in the research, and that Dan and I will be giving a webinar for them next month.

The webinar is called “When Teachers Talk Politics Online: The Dark Side of Online Spaces for Teacher Professional Learning” and will take place on October 6th at 4pm EST. It is free, but you need to register here to get the Zoom link.

404 Media podcast on generative AI and epistemology

Thu, 11 Sep 2025 09:00:30 -0400

I’m a big fan of the 404 Media tech news outlet, and I also really enjoy their podcast. I especially appreciated an episode that I listened to yesterday, which I’m embedding below as a YouTube video (as an aside, I simply do not understand how YouTube has become a major podcast-listening medium, so it pains me a bit to do this, but I’m once again trying to write something quickly before getting to real work, and YouTube embeds are relatively easy to do in Hugo, so that’s what I’m going with.

In the first segment of the podcast, Jason Koebler and Emmanuel Maiberg discuss a piece that the former wrote making the argument that “AI Generated ‘Boring History’ Videos are Flooding YouTube and Drowning Out Real History.” I hadn’t really read that piece, but as is often the case on the 404 Media podcast, I found myself appreciating their discussion of the piece even though I had skipped over it in text format.

There’s one point in particular that Koebler made that reminded me of some of the concerns I’ve been voicing recently about generative AI and epistemology. As journalists, I (genuinely!) don’t know whether Koebler and Maiberg think about epistemology and knowledge production in the same way that researchers do (or ought to). However, I was pleased by how squarely Koebler hit the nail on the head with how academic research works and how that is missing—or at least hidden—in (much? most? all?) generative AI output.

Here’s my sloppy copying from the YouTube transcript of the passage that really stood out to me, which begins at about 20:49 in the video above if you’d rather listen:

And then, you know, I’m not a historian. And I’ve watched a lot of things, but something I really appreciate is that history feels like it’s a conversation with different perspectives. And so a lot of these, the best humanmade channels will be like, “Well, this, this, like academic study or this historian says this, but like this other person who is also a well-renowned expert in the field says that, and it’s not quite exactly the same. Um, so you need to like consider it from these different perspectives and then sort of decide, you know, what you believe to be the truth. Um, and I think that there’s, there’s absolutely none of that in here. And it, it definitely doesn’t cite any its sources. Like it’s, it’s so funny. I’ll be listening to one of these Ancient Americas, uh, videos and it will be like, oh, like, the, these, uh, Harvard, uh, anthropologists wrote a study, you know, dating these ruins to a specific time, but like these other historians say that’s impossible because the migration patterns that, that were known during this time period don’t line up or whatever. And like, that’s, that’s obviously very vague, but, um, it’s a lot of that and I find that to be extremely interesting.

It’s time to get back to real work, so I’m not going to take the time to unpack everything that’s in there, but I don’t know that I need to—that’s pretty straightforward! While I know that there are potential and actual applications of generative AI that lay bare more of that epistemological work (and while there are more traditional sources of knowledge that also tend to conceal the epistemological work), I feel like this does a good job of getting at what my epistemological concerns about generative AI are.

new publication: online space in a Community of Christ congregation

Wed, 10 Sep 2025 13:20:34 -0400

A couple of years ago, I worked with a graduate student to examine the geographic and linguistic diversity in an online Community of Christ ministry that made an intentional effort to cross borders during the COVID-19 pandemic (and before and since, but for all of the obvious reasons, this was particularly pronounced during the pandemic). I was pleased with how the work went and was eager to get it to publication. I’m happy to announce that that’s finally happened, thanks to the Journal of Media and Religion.

There are a limited number of “eprints” available here, and I’ve uploaded a preprint here. This is also the first publication that I’ve been able to add to the new research section of my website as it came out. You can check out the publication page here or browse all of my research here—or via the taxonomies that I link to there. I’m pretty proud of that research section setup, so it probably deserves its own post sometime soon. For a simple preview of the article, here’s an abstract:

Internet technologies may provide new spaces for churches that are facing challenges in geographic spaces. These online spaces can be understood as distinct online spaces or as extensions of geographic spaces. We consider the Beyond the Walls online ministry provided by the Toronto Congregation of Community of Christ, a denomination with a growing global footprint and a shrinking population in the Global North. Examining worship services between January 2020 and January 2022, we consider the locations and languages represented in Beyond the Walls services, the distribution and social network of contributing individuals, and how the services performed on Facebook and YouTube. We found that online ministry allowed Beyond the Walls to draw from a larger, more geographically and linguistically diverse population than in-person services could have, possibly responding to denominational concerns. We also illustrate the ways that this ministry corresponds with both distinct and extended understandings of space in the online church.

🔗 linkblog: GOP megabill could cost UK HealthCare $100M annually in Medicaid reimbursement payments, but a leading university official has doubts

Tue, 09 Sep 2025 21:07:26 -0400

🔗 linkblog: GOP megabill could cost UK HealthCare $100M annually in Medicaid reimbursement payments, but a leading university official has doubts https://spencergreenhalgh.com/work/2025-09-09-the-effect/

🔗 linkblog: How Elon Musk Is Remaking Grok in His Image

Tue, 02 Sep 2025 07:00:22 -0400

Perhaps the best demonstration yet of why we need to talk about epistemology when we talk about generative AI. Gift link. It turns out that it takes an awful lot of intervention to get Grok to be “maximally truth-seeking” and “neutral.”

🔗 linkblog: Le Valais veut se doter d'une université prioritairement à distance

Mon, 01 Sep 2025 08:12:06 -0400

Je connaissais pas l’existence d’UniDistance Suisse, et maintenant j’ai envie d’en apprendre plus.

Thu, 28 Aug 2025 15:02:12 -0400

Today was really, truly going to be a much-needed writing day, and then I didn’t even clear my inbox until 2:30. Yayyyyyy.

Wed, 27 Aug 2025 17:24:16 -0400

Spaced on an evening meeting that I’d actually been looking forward to and 😫😫😫

🔗 linkblog: Former faculty leader leaves University of Kentucky after bucking president’s governance plan • Kentucky Lantern

Tue, 26 Aug 2025 18:55:17 -0400

I don’t have the whole picture here, and I don’t like passing judgment without having the whole picture, but I don’t know if I like this.

religious institutions, religious community, and religion-as-platform

Tue, 26 Aug 2025 17:39:06 -0400

I am very excited that Rosemary Avance is coming to one of the Mormon Social Science Association sessions at this year’s Society for the Scientific Study of Religion to speak on her book Mediated Mormons. I’ve just started the book in preparation for the session, and I was struck by the questions that make up the first two lines of the introduction:

What does it mean to be part of a religious community? Is it the same as claiming a religion?

I’m excited to see what Avance does with these questions, but I have to admit that my mind went first to a post I wrote last week, about what it would mean to treat religion theoretically as a platform. After all, online communities have a funny relationships with platforms. They may owe their existence to platforms, but they also exist somewhat independently of those platforms. Reddit provides a compelling example here: Subreddits obviously can’t exist without Reddit, but it’s the subreddits that do a lot of the heavy lifting. This tension becomes clear in moments like the Reddit blackout of Summer 2023, where subreddit moderators and communities clashed with platform executives in a quite remarkable way that I wish had had more of an impact on the platform’s direction.

I wonder if this kind of distinction between platform and community could be helpful in the context of “religion-as-platform.” For a number of reasons, I—along with many other scholars—have had to argue for making a distinction between “Mormonism” and “The Church of Jesus Christ of Latter-day Saints” in our research. Here, “Mormonism” is the community, and “The Church of Jesus Christ of Latter-day Saints” is the platform. The community owes its existence to the platform, but there’s a certain extent to which the community acts independently of the platform, and there are occasional tensions between the platform and the community. As with social media platforms, there are also explicit efforts by the platform to police the community.

This metaphor isn’t perfect, but I don’t think its imperfections are insurmountable, either. In some ways, Reddit has more control over subreddit communities than the Church of Jesus Christ of Latter-day Saints exercises over the Mormon community. That is, if you only understand the subreddit community as existing in that subreddit, a Reddit account ban is more powerful than a Latter-day Saint excommunication, which by no means prevents someone from engaging with the Mormon community. A sprinkling of James Paul Gee’s affinity spaces, though, might be helpful here, in which a space can transcend multiple platforms.

This is more spitballing, so don’t hold me to any of this yet, but I think there’s something to it—and I’m also excited to read more of Avance’s work instead of getting lost in my own tangents.

une série de France Culture sur Jacques Ellul

Tue, 26 Aug 2025 09:43:24 -0400

Merci à Matoo, qui a vu combien j’écrivais sur Jacques Ellul sur ce site et qui m’a donc recommandé la petite série de cinq épisodes « Avoir raison… avec Jacques Ellul », qui est sorti il y a quelques semaines sur France Culture. J’ai écouté la première épisode ce matin en faisant de petites préparations pour mon premier jour d’enseignement pour cette année scolaire, et je le trouve déjà très utile.

J’ai déjà lu trois livres par Ellul et je suis en train de lire deux autres (bon, en théorie — j’avoue que ça va lentement). Ce week-end, je vais recevoir quelques nouveaux livres d’Ellul que mon beau-frère a acheté à la librairie new-yorkaise magnifique Albertine, qui est soutenue par l’ambassade française aux États-Unis. Mon beau-frère va à New York tous les étés et me cherchent toujours quelques bouquins francophones, et ça fait qu’en ce moment, j’aurai bientôt beaucoup plus à lire d’Ellul. C’est la première fois dans ma vie que je m’engage à ce niveau avec l’œuvre d’un seul écrivain académique, et je trouve qu’avoir des résumés comme celui de France Culture m’aide beaucoup à situer ce que je lis en un moment particulier dans l’ensemble de sa pensée.

En plus, je me dis souvent que je devrais écouter plus de choses sur France Culture, donc merci encore à Matoo !

Quelques constats qui n’ont pas grand-chose à voir avec l’œuvre d’Ellul mais qui m’ont frappé lors de cette première épisode:

Je connaissais la tension qui existe entre Paris et « la province » en France, mais cela m’a quand-même surpris qu’on a suggéré que l’identité bordelaise d’Ellul a empêché ses écrits de recevoir plus d’attention en France lors de son vivant.
Je trouve qu’il y a un certain « accent français du 20ème siècle » que j’aime beaucoup, y compris chez Ellul. Je n’ai pas forcément l’oreille fine (même en anglais, ma langue maternelle), et je suis sûr que je rate quelques nuances, mais il y a une façon de parler et de prononcer que j’entends dans des enregistrements entre les années 40 et les années 90 qui m’intéresse beaucoup. C’est tout bête d’en écrire ici, mais entendre Ellul et sa façon de parler me plaît beaucoup. Comme francophile étranger, je m’intéresse toujours aux petites choses linguistiques qu’on nous a jamais apprises dans des manuels scolaires.

Tue, 26 Aug 2025 09:42:37 -0400

After years of putting this off, I have finally figured out how to network my computer to the department printer, and I no longer have to worry that my position as a tenured professor of information communication technology will be revoked for being intimidated by basic office setups.

🔗 linkblog: The NSF just cut K-12 STEM Education research going forward

Mon, 25 Aug 2025 11:44:18 -0400

🔗 linkblog: The NSF just cut K-12 STEM Education research going forward https://spencergreenhalgh.com/work/2025-08-25-appreciate-joshs/

Sat, 23 Aug 2025 17:36:02 -0400

Today, I attended my first kids’ soccer game since rereading James Carse’s “Finite and Infinite Games” and since reading about Jacques Ellul’s ethic of non-power, and it turns out that makes a difference?

defining platforms—and religion as platforms

Thu, 21 Aug 2025 21:00:23 -0400

I subscribe to the “Religion Watch” newsletter out of Baylor University but usually don’t do much more than skim it. The first entry in the June edition, though, immediately stood out to me for this excerpt:

Paul Seabright’s recent book, The Divine Economy: How Religions Compete for Wealth, Power, and People (Princeton University Press, $35), is unique for its comprehensive treatment of the religious past and present as well as its novel use of the concept of “platforms” in explaining the economy of religion.

As someone who studies religion on platforms, I was immediately intrigued by this idea of studying religion as platforms. I was lucky enough to be able to download the book through my institutional library, and I’ve been slowly making my way through it since. “Slowly” because I have a number of pressing deadlines—including the beginning of the semester next Monday—and as intrigued as I am with this idea, I really don’t have the research bandwidth right now to do something with this. However, I’ve read enough of Seabright’s book to have some initial thoughts on how he uses platform as a conceptual framework, and I’d like to write them down in case I have time to do something more substantive a few months from now.

Seabright on “platforms”

To be honest, I find Seabright’s understanding of platforms… disappointing. Now, as far as I’m concerned, theory is more rhetorical than ontological. That is—and to paraphrase Étienne Wenger—theory is useful for drawing the researcher’s attention to certain aspects of a phenomenon. I’m not interested in arguing about the necessary or sufficient features of “a platform,” and I’m not sure that one could do so even if one believed it was a useful exercise. I also concede that while I am an internet researcher, Seabright is an economist, and I am ill-suited for evaluating whether Seabright’s understanding of platforms draws the right kind of attention to the right aspects of religion from an economist’s point of view.

As that internet researcher, though, I’m not sure that Seabright really captures what makes platforms interesting, and I can’t see myself using his understanding of platforms to study religion. In fact, his repeated insistence that scholars’ focus on digital platforms distracts us from other manifestations of platforms, while interesting, also suggests to me an overly broad understanding of what platforms are. Here’s an early explanation, from p. 10 of his book:

Platforms are organizations that facilitate relationships that could not form, or could not function as effectively, in the platforms’ absence.

This is a fine definition, and if it draws attention to what Seabright is interested in, I think that’s theoretically sound enough for his purposes.

what makes digital platforms interesting

However, I don’t know that Seabright’s definition is sufficient to support an argument like this one, from p. 95:

Digital platforms such as Google, Twitter, and Instagram have become an ever-present feature of our online lives in recent years, to the point where we can exaggerate their novelty and overlook how important non-digital platforms have been for us in the past.

Just earlier this week, I was arguing for an expansive understanding of the term technology, one that helps us not exaggerate novelty, so I have a certain sympathy for this argument. Yet, as Tarleton Gillespie argues in his 2010 article The Politics of ‘Platforms’:

The point is not so much the word itself; ‘platform’ merely helps reveal the position that these intermediaries are trying to establish and the difficulty of doing so. YouTube must present its service not only to its users, but to advertisers, to major media producers it hopes to have as partners and to policymakers. The term ‘platform’ helps reveal how YouTube and others stage themselves for these constituencies, allowing them to make a broadly progressive sales pitch while also eliding the tensions inherent in their service: between user-generated and commercially-produced content, between cultivating community and serving up advertising, between intervening in the delivery of content and remaining neutral.

In short, “facilitate relationships” seems to me too timid a phrase to use to describe platforms, when digital platforms, as Gillespie suggests, invoke the phrase to suggest a neutrality despite the heavier hand that they actually play.

another example of a platform

Another thing I posted earlier this week was a link to an excellent (if depressing) 404 Media article about the Tea platform. I highly recommend the article, but let me give the briefest of summaries to illustrate what I think Seabright’s definition captures and what I think it misses.

Tea is a platform inspired by the ecosystem of “Are We Dating the Same Guy?” Facebook groups, which allow women to trade notes on men they’ve met through dating apps. So far, Seabright’s definition holds up: The point of Tea is to facilitate relationships between women so that they can share important information. Since other platforms facilitate other human behaviors besides relationships, let’s tweak Seabright’s phrase to facilitate human behavior.

Here’s where the Tea story gets wild, though. Tea was inspired by the “Are We Dating the Same Guy?” Facebook groups, but when the founder of those groups didn’t show any interest in joining the app team, Tea went to great lengths to infiltrate those groups, create deliberately similar alternative groups, and to badmouth those groups. So, Tea isn’t just interested in facilitating human behavior—it’s also interested in defining human behavior by shaping its boundaries and expressions. (In her The Culture of Connectivity, José van Dijck also provides a great example of how Facebook’s “Like” button defines—at least in that context—what it means to “like” something.)

And, of course, Tea is a Silicon Valley startup, so it’s trying to monetize its purportedly noble mission of protecting women in their dating experiences. Seabright is attentive to religion’s relationship with money—the subtitle of the book is “How Religions Compete for Wealth, Power, and People”—so it’s surprising to me that this isn’t built more explicitly into the platform model that he’s adopting. Let’s go ahead, though, and add profiting from human behavior as a third defining feature of a platform.

Now, this is some off-the-cuff thinking that mostly happened while driving back from kiddo’s soccer practice, so I’m not going to carve these three “platform attributes” into stone. I’d like to go back into the literature before setting up any definition of platform as a useful framework. Also, like any framework, there are some examples that fit better than others. Mastodon, for example, probably ought to count as a platform, but it’s hard to see how it “profits from human behavior” in the same way that Tea does.

applying this understanding to religion

Despite those caveats, though, I think this understanding of a platform as something that facilitates, defines, and profits from human behavior is much more compelling than the understanding that Seabright offers. What’s more, if the purpose of theory is to draw researchers’ attention to the most salient aspects of a phenomenon, this three-pronged approach strikes me as interesting for how it could draw attention to certain aspects of religion.

Let’s take the example of the Church of Jesus Christ of Latter-day Saints, since most of my religion research is related to Mormonism. Selecting “seeking salvation” as our human behavior—and setting aside the reality of any salvation-oriented cosmology for the purposes of this exercise—we could productively ask ourselves:

how does this church facilitate seeking salvation? (by providing access to certain sacraments and offering certain teachings)
how does this church define seeking salvation? (by adding necessary sacraments not found in other Christian faiths and adding a communal family element to salvation)
how does this church profit from seeking salvation? ( …even as a no-longer-practicing Latter-day Saint, I admit to finding this question a bit uncomfortable… and yet, there’s no denying that this is a very, very rich church)

As an internet researcher, I would be further interested in asking how these platform logics might influence the way that Latter-day Saint institutions employ, avoid, or emphasize other technological platforms.

conclusion

This is a lot of spitballing, and I really ought to be getting to bed, so spitballing it will stay. I think Seabright is onto something by considering religions as platforms, but I don’t think he offers the right understanding of “platforms” for that kind of consideration to be useful to me and my work. That said, if this spitballing has accomplished one thing, I hope it’s to demonstrate that there are interesting characteristics of platforms—and that a more developed understanding of those characteristics could lend itself productively to research involving religion.

🔗 linkblog: How Tea’s Founder Convinced Millions of Women to Spill Their Secrets, Then Exposed Them to the World

Tue, 19 Aug 2025 19:46:00 -0400

What a wild, depressing story. I feel like I ought to use this to teach the concept of platforms to my students—it neatly sums up the intervention in normal human activity by someone who thinks they have a buck to make.

🔗 linkblog: An AI divide is growing in schools. This camp wants to level the playing field

Tue, 19 Aug 2025 06:49:10 -0400

Closing digital divides is good, and increasing diversity in tech fields as good, but I’ve been complaining for years about computer science ed that we stop at the nobility of those goals and don’t ask ourselves about the deeper motivations behind those initiatives. So it is with AI: A more diverse field more available to all is better than what we have, but we also have to ask whether AI education is actually a social good.

insisting that pencils are technology is not (necessarily) a wiseass move

Mon, 18 Aug 2025 15:52:01 -0400

Thanks to the magic of Bluesky, I came across Paul Musgrave’s essay “Classroom Technology Was a Mistake,” with the subtitle “Hopes that AI will improve higher ed need to reckon with the dashed hopes of the past.” As a whole, I appreciate the essay—I’m sympathetic to Musgrave’s argument, and I couldn’t agree with the subtitle more if I tried. I want to do one of those things, though, where one academic spends too much time quibbling with a minor part of another academic’s argument. In particular, I want to take issue with this part of Musgrave’s essay:

Now, there’s always some wiseass scholar in the audience who wants to say that “technology has always been in the classroom” and they want to talk about how pens or pencils or flint arrows are “technology”, but we all know that what we mean are computers—the most powerful general-purpose tool of the past one hundred years.

In this post, I am going to be that wiseass for a second, but as the title of this post suggests, while I’m sure there are people who invoke pencils as technology to be the wiseass who dismisses Musgrave’s concerns about educational technologies, I think there’s a productive way to understand pencils as technology, too. In fact, I think Musgrave’s skepticism about the purported revolutionary potential of Canvas and AI gets even stronger if we broaden “technology” beyond the general-purpose computer.

Musgrave begins his essay with some reflection on the “Oregon Trail generation” (born between 1977 and 1983) being the “first for which ’technology in the classroom’ was a guiding principle of our educators.” Consider, though, that in 1986, when even the oldest children of that generation were still in primary school, educational technology critic Larry Cuban was already writing in Teachers and Machines: The Classroom Use of Technology Since 1920 about the failures of radio and television to deliver on their purported revolutionary potential and drawing lines between those technologies and the computer. Audrey Watters, in the much more recent Teaching Machines: The History of Personalized Learning, traced many of the modern problems with educational technologies back to B. F. Skinner’s work in the early 20th century. Hell, I enjoy showing students this blurb from Josiah F. Bumstead’s (couldn’t make up a better name for this example if I tried) 1841 book The black board in the primary school:

The inventor or introducer of the black-board system deserves to be ranked among the best contributors to learning and science, if not among the greatest benefactors of mankind; and so he will be regarded by all who know its merits, and are familiar with school-room trials.

As Bumstead demonstrates, misplaced hype about educational technology is nothing new; as Cuban demonstrates, the problems with that misplaced hype are nothing new. In fact, let’s go to an even broader (but arguably even less wiseass) definition of what (educational) technology is to show how a broad definition of technology can actually give us the very conceptual tools we need to critique it. Over the course of 2025, I’ve been reading a lot of Jacques Ellul, and I think that he provides a helpful definition of technique that we can use (in a scholarly rather than wiseass fashion) to make sense of both pencils and generative AI as belonging to the same category and yet qualitatively different. In his The Technological Society, Ellul defines technique as follows:

“The term technique, as I use it, does not mean machines, technology, or this or that procedure for attaining an end. In our technological society, technique is the totality of methods rationally arrived at and having absolute efficiency (for a given stage of development) in every field of human activity.

Ellul’s definition of technique (which we’re going to consider here as equivalent to technology, even though he would hate that) is so broad as to almost be useless. Yet, his marrying of technique to efficiency is helpful here. The push for Canvas and generative AI in the college classroom comes from the same fetishization of efficiency that Ellul was concerned about in the 1950s. And yet, to quote Musgrave:

That digitalization looks modern and efficient but it often is anything but. It takes me hours longer to create a course shell than it used to and the results are not appreciably better; indeed, because of shortcomings in the tools, I find myself frequently battling between adapting my course to the machine or bashing the machine into doing what I want.

Ellul’s loose definition of technique as that which is adopted or imposed in the name of efficiency (and efficacy) fits nicely with Musgrave’s (and countless others’) observation that that which is adopted or imposed in the name of efficiency creates new problems that we haven’t fully thought through. Let’s even take this a step further: Ellul’s emphasis on efficacy and efficiency may draw our attention to other roots of the problem that go beyond “computers” or “software.” Here’s the French sociologist’s skepticism of government funding of research, in a way that feels painfully familiar in 2025:

The state claims to represent the public interest and hence to have the duty of being a “good manager,” dispensing the public revenues only on condition that they mean something, that they pay off.

There’s no denying that the Trump Administration is acting in bad faith in cutting research funding “in the public interest,” and yet Ellul’s broad definition of technology helps us understand it in the same vein that we understand its relentless cheerleading of generative AI in schools, and countless other appeals to efficiency that are societally detrimental.

I’ve spent longer on this post than Musgrave’s self-imposed 20 minutes, and it’s time for me to go back to my own wrestling with Canvas to get it to do what I want it to do. I appreciate the original argument of the post, and I’ll be thinking about it in the months to come—but I hope I’ve demonstrated how that argument can get richer when we accept the pencil as another educational technology.

🔗 linkblog: What's behind the Trump administration's immigration memes?

Mon, 18 Aug 2025 06:59:57 -0400

There’s always been a dark side to internet culture, but I don’t think it was naïve in my earlier work to argue for recognizing its value. Yet, it’s important as a scholar to call out the ugliness that’s happening here.

why I think labor, not copyright, is the foundational problem with AI scrapers

Fri, 15 Aug 2025 11:33:52 -0400

This morning on Bluesky, I saw some posts about a class action lawsuit against Anthropic for their use of pirated, copyrighted materials in training their generative AI models. One of the sources of these copyrighted materials was the LibGen database, which I took a peek at nearly six months ago to confirm what I was already sure to be true: that my scientific writing was also collected as training material by companies like Anthropic or Meta. I don’t love that big tech companies are profiting off of my work in this way, and I’m sympathetic to the authors who are taking legal action against Anthropic. However, as I’ve written repeatedly over the past few years (you can find some of those thoughts—and others—by scrolling through here, I don’t know that copyright is the right way of responding to this kind of abuse.

As a scientific—rather than literary—author, I recognize that I have a different perspective on this issue, and I’m not interested in telling authors that they’re making the wrong move by joining this class action lawsuit. However, as a scientific—rather than literary—author, I think my experience can also help illustrate why I don’t think that copyright is the foundational issue here. For all my hesitation about copyright as the best vector of criticism toward content scraping by generative AI companies, I did entertain the idea this morning that it might be interesting to join myself to a class action lawsuit. While skimming the page, though, I noticed something interesting in the very first sentence (my emphasis added):

Bartz v. Anthropic PBC is a class action lawsuit under the Copyright Act brought by authors on behalf of copyright holders against Anthropic PBC, an AI company.

The phrase “copyright holders” reminded me of a depressing quirk in scientific publishing: That I actually do not hold the copyright to a great deal of what I’ve written. Despite all of the work that I’ve put into this kind of writing since 2013, I don’t always belong to the class being represented in this lawsuit. In fact, I went back through The Atlantic’s database of LibGen data to check this and confirmed that many (if not all—I didn’t take the time to be totally thorough) of my articles that Meta and Anthropic presumably used as training data don’t legally belong to me or to any of my co-authors.

On a regular basis, I have to sign over my copyright to a scientific publisher before they will agree to publish my work, and that makes them the ones who have been mistreated according to a copyright framing of what (to be very clear) is still a really big problem. To use an analogy from Rebecca Giblin and Cory Doctorow’s Chokepoint Capitalism:

giving more copyright to creators who are struggling against powerful buyers is like giving more lunch money to your bullied kid. The bullies who were taking his money every day will just take that too. The upshot? The bullies now have enough money to pay the principal to look the other way, and your kid still goes hungry.

In short, scientific authors are already being exploited for their academic labor by scientific publishing companies, who often claim the copyright (or the distribution in the cases where they don’t claim copyright) to our work. Scientific publishers make far more money from our work than we ever will. To make copyright the tool by which we defend authors against exploitation by generative AI companies may actually support and further the exploitation of scientific authors by their publishing companies.

Again, I recognize that things are different for literary authors, and I also acknowledge that I don’t know of any legal mechanisms to protect authors’ labor in the same way that intellectual property law allows for the protection of authors’ copyright. Yet, it seems clear to me that in among both communities, it is labor that is being exploited, and that we need to fix the problem of labor exploitation (and not just intellectual property) if we’re really going to fix this issue.

🔗 linkblog: Google Scholar Is Doomed

Wed, 13 Aug 2025 06:41:24 -0400

Oof, hadn’t thought of this, but as much as I’d like to further reduce Google dependence, this would really hurt.

🔗 linkblog: The Trump Administration Is Using Memes to Turn Mass Deportation Into One Big Joke

Tue, 12 Aug 2025 14:17:25 -0400

🔗 linkblog: The Trump Administration Is Using Memes to Turn Mass Deportation Into One Big Joke https://spencergreenhalgh.com/work/2025-08-12-bookmarking-this/

🔗 linkblog: Phénomène mondial sur les réseaux sociaux, que sont les Italian Brainrots, ces personnages absurdes générés par IA ?

Mon, 11 Aug 2025 16:33:58 -0400

Ma carrière se divise entre une valorisation des pratiques numériques perçues comme n’ayant pas d’importance et une critique des technologies qui permettent ces pratiques. J’avoue que je me sens vraiment divisé pour cet exemple !

Sun, 10 Aug 2025 08:49:08 -0400

Je viens d’entendre à la RTS une définition des « digital natives » comme la génération Z. On a commencé à employer cette phrase aux années 2000 pour décrire ma génération, et son usage continu pour signifier « les jeunes d’aujourd’hui » révèle combien elle est inutile.

Fri, 08 Aug 2025 20:23:31 -0400

Love that moment in academic writing where you can find five rejected versions of a paragraph below the version I’m currently working on.

🔗 linkblog: New executive order puts all grants under political control

Fri, 08 Aug 2025 07:09:42 -0400

Here’s Jacques Ellul on state funding of research:

The state demands that anything scientific enter into the line of “normal” development, not only for the stake of the public interest but also because of its will to power. We have previously noted that this will to power has found in technique an extraordinary means of expression. The state quickly comes to demand that technique keep its promises and be an effective servant of state power. Everything not of direct interest to this drive for power appears valueless.

I think Ellul’s observation was true even under the old U.S. regime of state funding, but wow is it ever true now.

🔗 linkblog: Nouveau modèle d'IA: ChatGPT-5: «C’est comme parler à un expert de niveau doctorat»

Fri, 08 Aug 2025 07:06:13 -0400

À mon avis, l’expertise « au niveau doctorat » est surtout lié au processus de connaître et non à la connaissance elle-même, et l’IA générative ne respecte pas du tout ce processus.

🔗 linkblog: Google would like you to study with Gemini instead of cheat with it

Wed, 06 Aug 2025 17:31:43 -0400

This seems performative to me, and this paragraph gets at why I think so:

AI companies are increasingly pushing into education — perhaps in part to try and fight the reputation that AI tools have acquired that they help students cheat. Features like Gemini’s guided learning mode and ChatGPT’s similar study mode, which was announced last week, could theoretically help with actual learning, but the question is whether students will want to use these modes instead of just using the AI chatbots for easy answers.

So, first, Google and OpenAI (and everyone else) want to get into education because it’s profitable. They can get lots of money from university partnerships and even more from winning over future customers. Second, that last question is really important: Will the students using generative AI switch to a study mode? Or is this a fig leaf that allows these companies to demonstrate how educational they are so that they can make those profits?

🔗 linkblog: Is Including Hidden AI Prompts In Academic Papers Gaming The Peer Review System — Or Keeping It Honest?

Sat, 26 Jul 2025 07:44:10 -0400

Kind of hate that we have to ask the question in this headline!

I’ve been (link)blogging recently about needing to talk about epistemology when we talk about generative AI. I know that in at least one case, I’ve evoked the generation of scientific knowledge as a counterexample to the “just the facts, ma’am” naïve epistemology promoted by AI and its supporters. To use generative AI to review papers—and to try to get around peer review—feels particularly dangerous to me.

Thu, 24 Jul 2025 20:23:04 -0400

I’m glad I have my files organized enough to retrieve preprints of about a decade’s worth of research, but I really wish I’d been doing this as I went.

🔗 linkblog: The White House orders tech companies to make AI bigoted again

Thu, 24 Jul 2025 18:38:17 -0400

Quick question about this passage:

Trump … signed an executive order titled “Preventing Woke AI in the Federal Government,” directing government agencies “not to procure models that sacrifice truthfulness and accuracy to ideological agendas.”

So, how does one determine what is true or accurate? Once again, we need to talk about epistemology when we talk about generative AI.

🔗 linkblog: Instructure and OpenAI Announce Global Partnership to Embed AI Learning Experiences within Canvas

Thu, 24 Jul 2025 09:45:50 -0400

🔗 linkblog: Instructure and OpenAI Announce Global Partnership to Embed AI Learning Experiences within Canvas https://spencergreenhalgh.com/work/2025-07-24-dammit-am/

🔗 linkblog: Trump unveils his plan to put AI in everything

Wed, 23 Jul 2025 13:42:06 -0400

This emphasis on “objective truth” further underscores the need to talk epistemology when we talk AI.

🔗 linkblog: Une lycéenne accusée d'avoir triché avec une IA au baccalauréat de philosophie obtient finalement son diplôme

Wed, 23 Jul 2025 13:40:14 -0400

🔗 linkblog: Une lycéenne accusée d'avoir triché avec une IA au baccalauréat de philosophie obtient finalement son diplôme https://spencergreenhalgh.com/work/2025-07-23-je-naime/

« en présentiel » et d'autres phrases à apprendre pour une traduction de CV

Tue, 22 Jul 2025 07:12:37 -0400

Ce weekend, quelqu’un m’a recommandé une épisode de la série « La science et ses mauvaises consciences » , qui fait partie de l’émission Avec philosophie sur France Culture. J’ai décidé de télécharger toute la série, et en écoutant la première épisode, j’ai entendu une des interlocutrices se servir de la phrase « en présentiel », après quoi elle s’est excusée pour avoir prononcé un anglicisme.

Cela m’a gêné un peu, car ça faisait quelques jours que je travaillais sur une version de mon CV en français, une partie importante de mes efforts d’avoir un site web plus ou moins bilingue. Comme je suis professeur, l’enseignement fait évidemment partie de mon CV. Je fais beaucoup d’enseignement en ligne, et je m’étais donc servi de la phrase « en présentiel » pour distinguer les autres cours qui se déroulent dans des salles de classe. Est-ce que j’avais fait une faute ?

Heureusement, je croyais me souvenir que le bouquin Le français va très bien, merci (que j’avais acheté à Colmar lors de mes vacances hivernales) avait mentionné cette phrase-là, et j’ai donc feuilleté le livre pour retrouver le passage en question. Voici ce qui Les linguistes attéré(e)s en disent :

En France et en Afrique francophone, l’Académie française bénéficie d’une aura symbolique, d’un prestige que lui confèrent son ancienneté, son apparat (bâtiments, costumes, épées), ses membres (écrivains reconnus, politiques…). On y accède sans aucune formation à la linguistique. Le travail sur le dictionnaire est réalisé par des agrégés de lettres dont le public ignore l’identité. Il en est de même pour sa rubrique «Dire, ne pas dire», mise à jour de manière très inégale. Elle a ainsi préconisé en juillet 2020 de ne pas utiliser les mots distanciel et présentiel sous prétexte qu’ils seraient des anglicismes. Ce qu’ils ne sont pas…

Comme le confirme cette dernière phrase, le mot « présentiel » n’est pas un anglicisme. En fait, quand j’ai lu ce bouquin pour la première fois, l’idée que cette phrase puisse être critiquée comme anglicisme m’a choqué, car je ne connaissais pas la phrase équivalente en anglais. (On a plutôt tendance à dire « in person » où bien « face-to-face »). Il paraît qu’il y a un mot anglais « presential », mais je ne l’ai jamais entendu de ma vie. Ce n’est qu’en recherchant l’étymologie de la phrase française que j’ai découvert cette racine anglophone.

Malgré ma confiance renouvelée, je me demande toujours si celleux qui liront la traduction de mon CV vont croire (tout comme l’animatrice à France Culture) que j’ai fait le fainéant américain en choisissant un anglicisme pour décrire certains cours, et c’est en fait ce petit souci qui a inspiré ce post de blog.

En tout cas, même si je peux défendre le choix d’écrire « en présentiel », j’avoue que c’était vraiment difficile de faire cette traduction d’une CV. Comme est souvent le cas, la difficulté vient du fait qu’au-delà des différences de langue, il y a des différences de conception entre les deux cultures.

Dans ce cas-ci, on peut dire sans trop d’hésitation les universités en Francophonie sont différentes que celles aux États-Unis. Par example, alors qu’en Europe (y compris l’Europe francophone), on distingue entre trois cycles universitaires, aux États-Unis, on fait plutôt distinction entre « undergraduate » (Bachelor, c’est-à-dire, le premier cycle) et « graduate » (Master et Doctorat, c’est-à-dire le deuxième et le troisième cycle ensemble).

Comment donc préciser que je suis reconnu comme « graduate faculty » à mon université ? Est-ce que je suis « professeur du deuxième et du troisième cycle » ? C’est vachement compliqué, cela. Heureusement, les universités canadiennes ressemblent plus aux universités étatsunisiennes, et j’ai pu donc trouver la phrase « études supérieures » comme équivalent de « graduate » en anglais. Je suis donc membre de la faculté des études supérieures, et il arrive aussi que j’ai deux certificats des études supérieures (en plus de mes diplômes).

Cette solution n’est pas parfaite, car j’ai maintenant une mélange des français sur mon CV, m’étant inspiré et du français canadien et du français européen pour arriver à tout expliquer. J’espère que les lecteurs éventuels de ce document me pardonneront ce bricolage linguistique, mais je ne vois pas d’autre solution. Malgré toutes ces difficultés, il a été bien intéressant de faire cette traduction, et j’ai appris beaucoup de vocabulaire académique dont j’espère pouvoir me servir dans l’avenir. En plus, si quelqu’un critique mon usage de la phrase « en présentiel », je saurai me défendre.

🔗 linkblog: The NIH Is Capping Research Proposals Because It's Overwhelmed by AI Submissions

Tue, 22 Jul 2025 06:29:51 -0400

🔗 linkblog: The NIH Is Capping Research Proposals Because It's Overwhelmed by AI Submissions https://spencergreenhalgh.com/work/2025-07-22-this-sounds/

Mon, 21 Jul 2025 15:26:31 -0400

Just sent one of the most stressful emails of my career and got an out of office reply. I’m not mad—good for people taking time away from work—but that’s a heck of an anticlimax.

🔗 linkblog: Why I’m not going back to the U.S.

Fri, 18 Jul 2025 11:58:05 -0400

🔗 linkblog: Why I’m not going back to the U.S. https://spencergreenhalgh.com/work/2025-07-18-this-is/

🔗 linkblog: The Em Dash Responds to the AI Allegations

Thu, 17 Jul 2025 12:53:03 -0400

🔗 linkblog: The Em Dash Responds to the AI Allegations https://spencergreenhalgh.com/work/2025-07-17-as-a/

Thu, 17 Jul 2025 10:29:48 -0400

Je compte bientôt préparer un document professionnel en français. Il y a cinq ans, j’aurais demandé à quelqu’un de m’aider à corriger des erreurs. En 2025, par contre, je me demande si mes petites erreurs éventuelles feront preuve que j’ai tout écrit moi-même au lieu de me fier à l’IA.

🔗 linkblog: 'Arbres', 'noix', 'tout le monde sait'... Ce vocabulaire détourné par des internautes pour diffuser des idées d'extrême droite sur les réseaux sociaux

Thu, 17 Jul 2025 08:02:13 -0400

🔗 linkblog: 'Arbres', 'noix', 'tout le monde sait'... Ce vocabulaire détourné par des internautes pour diffuser des idées d'extrême droite sur les réseaux sociaux https://spencergreenhalgh.com/work/2025-07-17-le-d%C3%A9cryptage/

🔗 linkblog: 'Doing their own thing': KY legislators hear about the state of AI use and guidance in schools

Wed, 16 Jul 2025 08:04:29 -0400

I can see the value in some state guidelines, but I suspect they would be more permissive than what I want for my classroom. I hope I’ll still have the chance to establish restrictions as I see fit.

🔗 linkblog: Kentucky Republican lawmaker questions gender and women’s studies course at UK • Kentucky Lantern

Wed, 16 Jul 2025 08:01:24 -0400

So far, we’ve been told that the General Assembly’s war on DEI doesn’t affect our classroom teaching. Is this legislative bluster, or do we have worries on the horizon?

📚 bookblog: Mormons, Musical Theater, and Belonging in America (❤️❤️❤️❤️🖤)

Sun, 13 Jul 2025 13:42:51 -0400

I mostly skimmed this book, and I would have some quibbles with it if I got more into the details, but I found it really good. Musical theater is far, faaaar outside of my research interests, but this book articulates a fascinating “theology of voice” within Mormonism that will be helpful as I look to write something on Ellul and Mormon Studies.

🔗 linkblog: Grok searches for Elon Musk’s opinion before answering tough questions

Fri, 11 Jul 2025 08:23:04 -0400

🔗 linkblog: Grok searches for Elon Musk’s opinion before answering tough questions https://spencergreenhalgh.com/work/2025-07-11-look-i/

🔗 linkblog: Musk’s Grok 4 launches one day after chatbot generated Hitler praise on X

Thu, 10 Jul 2025 14:15:57 -0400

Okay, really don’t want to spend any more time writing about Grok, but let’s talk about this passage:

“With respect to academic questions, Grok 4 is better than PhD level in every subject, no exceptions,” Musk claimed during the livestream. We’ve previously covered nebulous claims about “PhD-level” AI, finding them to be generally specious marketing talk.

To return to my thoughts on AI and epistemology, I don’t think having a PhD is (or should be) a benchmark for content knowledge. Rather, I think it demonstrates (or should demonstrate) a commitment to the process of knowledge production, and LLMs cannot truly compete with humans there.

🔗 linkblog: Trump Seeks to Cut Basic Scientific Research by Roughly One-Third, Report Shows

Thu, 10 Jul 2025 11:32:15 -0400

Reading this through an Ellulian lens is interesting. In the 1950s, he was expressing concern about the valuing of (applied) technique over (basic) science. In this article, though, it’s clear how often that basic science is still described and defended in applied/technical terms. pushing the boundaries of knowledge seems to only be valuable if it “sow[s] practical spinoffs and breakthroughs” or helps the U.S. in its geopolitical competition.

Gift link.

🔗 linkblog: Musk makes grand promises about Grok 4 in the wake of a Nazi chatbot meltdown

Thu, 10 Jul 2025 07:21:09 -0400

Yesterday, I wrote my thoughts on how Grok’s “Nazi meltdown” helps illustrate some of my concerns about AI and epistemology.

This coverage of Grok’s latest demo only reinforces that—Musk’s tinkering with the LLM to get the results he wants is at odds with his states naïve epistemology that an LLM can be “maximally truth-seeking,” as though there is a self-evident truth that an LLM can deliver in a straightforward way (that is, without all that mucking about behind the scenes).

I’m also reminded of Karen Hao’s excellent Empire of AI, where she contrasts the lofty dreams (“discover new physics”) and fears (e.g., rogue superintelligence) of AI companies with their general refusal to confront the much more mundane problems these technologies create (“antisemitic tirades,” as the article puts it.

on Grok, other LLMs, and epistemology

Wed, 09 Jul 2025 15:44:15 -0400

Yesterday, I blogged (en français) on Jacques Ellul’s emphasis on the need for a technology-responsive ethic that emphasizes (among other values) tension and conflict. Ellul explores this ethic—one of non-power—in a few different writings that feel like different drafts of the same thing, and so I’ve seen that emphasis come up a few times as I’ve tried reading his work. Every time, it surprises me a little bit. Why, in articulating an ethical framework, would you emphasize tension and conflict?

Once you take in the context, this makes a lot more sense. Ellul’s concern is that technique—for the purpose of this blog post, technology, even though it bothers him for the term to be translated that way—tends to encourage and demand unity. In contrast, he argues, human society must disagree, debate, and discuss in order to survive. I already wrote some yesterday about how I think this observation is useful in the context of LLMs, but Elon Musk’s Grok has been in the news recently for worrying reasons, and so I want to develop this idea a little bit more and connect it to epistemology: a recurring concern of mine related to LLMs.

some Grok history

First, let’s dig into some historical context for Grok. Back in 2023, the Associated Press reported (based on a Fox News interview between Musk and Carlson) that the billionaire was looking:

to create an alternative to the popular AI chatbot ChatGPT that he is calling “TruthGPT,” which will be a “maximum truth-seeking AI that tries to understand the nature of the universe.”

Musk’s interest in LLMs betrays a certain epistemological stance: That there exists (implicitly self-evident) “truth,” or “nature of the universe.” While Musk’s dreams for a “TruthGPT” are built on an argument that not all LLMs are concerned with this truth, they are also clearly built on a similar foundation as Ellul’s concerns: That a technology can be designed in a way to provide and promote this truth in a way that cannot be argued against. Even if the billionaire’s efforts can be seen as entering into tension and conflict with the supposed liberal bias of other LLMs, it seems likely that he’d like to see unification around truth as understood by him and Grok.

Even if the product didn’t end up with the “TruthGPT” name, Musk’s epistemological commitments seem to be an influential part of how Grok works. A couple of months ago, in response to controversy about Grok’s tendency to bring up “white genocide,” Musk’s xAI released the system-level prompts for the LLM, which, according to The Verge, include references to “truth-seeking” as a “core belief” and instruction to “provide truthful… insights.”

shaky epistemological foundations

Here’s the thing, though. It sounds great to appeal to truth. It’s hard to argue against truth. However, truth isn’t as self-evident as Musk’s epistemology claims it to be. (To be fair, I spent large parts of my life with a similar epistemological commitment to truth, and the argument I’m making now would have been unwelcome and uncomfortable for me in my early-to-mid 20s. While I’m parenthesizing, I should also emphasize that I am not using the term truth as Ellul does in his writings.)

I’m not arguing that truth doesn’t exist, just that it’s actually quite difficult to determine and access. To take an example from a qualitative research methods book I used in grad school, either anthropocentric climate change is happening or it is not. That’s a question of truth. However, as firmly as I accept the widespread scientific consensus that anthropocentric climate change is truly happening, with great consequences for our species and our planet, I also think it’s important to acknowledge that the reality of this man-made disaster is… not straightforward.

Rather, it is a product of huge amounts of scientific research making elaborate arguments based on complicated statistical models, and the same “lack of another earth” that makes it so damn important to respond effectively to anthropocentric climate change also robs us of the kind of experimental methods that scientists would really prefer to use to establish the reality of this kind of thing. This is a very important truth, but it’s not the kind of casually self-evident truth that Musk believes an LLM can produce.

In fact, we could argue that it is through the tensions and conflicts that Ellul values that we arrived at the scientific consensus around anthropocentric climate change. I’d go even further to say that everything that we know as true—even the most seemingly self-evident mathematical and scientific truths—can only be achieved through tension and conflict. In my view as a (social) scientist, if we want to talk about truth, we have to talk about the processes by which we argue and determine what truth is.

Musk’s entrance into conflict

Furthermore, as I suggested earlier, the release of Grok can be seen as “entering into tension and conflict” with other LLMs. According to the AP, Musk’s dream of TruthGPT was driven in part by a concern that LLMs were being trained to be “politically correct.” However, his response does not appear to have been a hands-off approach to seek truth wherever it takes things, but rather to intentionally push in the other direction. I earlier quoted The Verge’s excerpt of Grok system prompts as “provide truthful… insights,” but I intentionally left out a couple of words there for dramatic purposes. The full quote is, with my emphasis: “provide truthful and based insights.”

That brings me around to the news stories that—alongside my reading of Ellul—inspired this post. Let’s go once again to an article from The Verge:

On Tuesday, X users observed Grok celebrating Adolf Hitler and making antisemitic posts, and X owner xAI now says it’s “actively working to remove” what it calls “inappropriate posts” made by the AI chatbot. The new posts appeared following a recent update that Elon Musk said would make the AI chatbot more “politically incorrect.” Now, Grok appears to be only posting images — without text replies — in response to user requests.

In short, it looks like Musk’s distaste for “political correctness” has led him to put his finger on the scale of how Grok operates (and, presumably, understands truth). Grok has been controversial a number of times in the past, but this really seems to have crossed a line, with the chatbot explicitly praising Hitler and making the company backtrack on its emphasis on political incorrectness.

This is horrifying and disgusting, but let’s focus for a bit on the way that it’s clear that even with his articulation of a naïve epistemology of “truth and nothing but,” Musk himself (through his employees) has to wade into tension and conflict (even with his own LLM) to obtain the truth that he’s looking for. I don’t believe that Musk is engaging in conflict and tension over truth in the good faith way that the scientific community typically does, but it is a useful example that even those who appeal to this kind of straightforward, self-evident truth tend not to find it when pressed.

broader implications for LLMs—and educational uses of them

I’ve been wanting to write about generative AI and epistemology for ages, and in this last section, I want to get at why. Musk and Grok serve as a particularly dramatic example, but they illustrate some important points about LLMs that I think deserve more attention. First, that they seem to provide a straightforward, self-evident truth for people to unify around. Sure, ChatGPT isn’t the obvious dumpster fire that Grok is, but there’s still a similarly naïve epistemology behind how it’s designed and how it’s used. Second, that despite the seeming straightforward nature of an LLM chatbot’s responses, there’s tension and conflict going on behind the scenes to resolve what correct answers are.

I think this second point is particularly important because of a conversation that I had with a friend who’s keener on AI than I am shortly after ChatGPT was released. He compared GPT-type tools to Wikipedia in terms of our distrust of their reliability as compared to sources that we’re more familiar with. I understand my friend’s point, but in the years since the conversation, this comparison has rankled me. While Wikipedia can—and does—get things wrong, it actually does quite a good job of baring its epistemological foundations to the world. You can go through edit histories. You can read up on editor discussions and debates. You can get a sense of how the article got to the point that it did. You have to know to look for those things, of course, but they’re there. My impression is that GPT-type chatbots do not do this. They provide an answer, and do not do as much work to lay bare the conflicts and tensions that lead to that answer.

It seems to me that enthusiasm about LLMs as an educational technology is based on its ability to provide students with easy access to facts and knowledge—truth, if you will. However, that seems to me to be a variation of the same naïve epistemology that I’ve been complaining about. Some of the most interesting and progressive approaches to pedagogy and educational technology over the past century(!) have emphasized engaging students in the conflicts and tension that are part of discussing, debating, and determining truth. Good history classrooms teach students to think like historians. Good science classrooms teach students to think like scientists. This is a much more robust epistemology, one that concedes that truth takes some work to get to and invites learners to engage in that work. Why not do more of that instead of retreat to the naïve epistemology of LLMs?

conclusion

Like so many blog posts, this is a jumble of thoughts that have been on my mind and could probably use more polish. I’m not a philosopher, so some of my thinking about epistemology is probably naïve at a meta-level. I don’t have any hands-on experience with LLMs, so I’d be happy for any misunderstandings there to be corrected. For all its roughness, though, this post gets at something that’s been nagging at me for months and something that I think needs a lot more attention. I hope to elaborate on this thinking more in the future.

Tue, 08 Jul 2025 13:33:48 -0400

Dealing with yet another case of “these reviewer suggestions are good and pretty straightforward, so why don’t I have any motivation to make the revisions?”

🔗 linkblog: Microsoft, OpenAI, and a US Teachers’ Union Are Hatching a Plan to ‘Bring AI into the Classroom’

Tue, 08 Jul 2025 11:57:58 -0400

🔗 linkblog: Microsoft, OpenAI, and a US Teachers’ Union Are Hatching a Plan to ‘Bring AI into the Classroom’ https://spencergreenhalgh.com/work/2025-07-08-it-feels/

Jacques Ellul contre l'IA

Tue, 08 Jul 2025 10:59:48 -0400

Ça fait plusieurs mois que je m’intéresse aux écrits de Jacques Ellul comme base théorique pour comprendre les techniques et technologies de nos jours. En fait, j’ai déjà écrit en février au sujet de l’intelligence artificielle générative et combien l’œuvre d’Ellul semble utile pour les critiques de l’IA malgré le fait qu’Ellul a vécu et écrit bien avant l’ère de l’IA comme nous la connaissons aujourd’hui.

Je suis en train de lire son livre posthume Théologie et technique (bien lentement, il faut l’avouer—j’avais commencé le livre en mai avant de devoir recommencer il y a quelques jours), et je trouve qu’il y a plusieurs passages qui me semblent utile lors des débats actuels au sujet de l’IA générative.

la nécessité des tensions et des conflits

Premièrement, aux pages 68-69 du livre, Ellul parle de son éthique de la non-puissance, une éthique qui’il décrit comme « créatrice de tensions et de conflits. » Il donne même l’exemple suivant :

Une société humaine n’existe que par des négociations successives de position contradictoires. Or, par exemple, la décision assistée par ordinateur exclut la Négociation.

Hier, j’ai posté un lien vers cet article de France Info sur la transformation des moteurs de recherche en « moteurs de réponse ». En voici un extrait (que j’ai déjà cité dans mon « linkpost » :

On est passé d’un moteur de recherche à un moteur de réponse. C’est-à-dire que les algorithmes proposent des versions rédigées à partir des données qu’ils auront collectées sur Internet, puis reformulées sans que vous ayez rendu visite aux sites contenant ces éléments de réponse à votre requête.

C’est sûr qu’on a déjà des problèmes à enseigner aux jeunes la nécessité « d’entrer en conflit » avec ces sites et ses éléments de réponse, mais un moteur de recherche présuppose (d’une certaine manière) que la personne qui cherche une réponse aura cette opportunité de mettre ces éléments en tension les uns avec les autres pour ensuite trouver une réponse par la négociation. Par contre, un moteur de réponse insiste sur « une concordance, une unité » (pour encore citer Ellul) en présupposant que c’est la technique déjà qui connaît la réponse, et les personnes n’ont plus rien à faire quant à la négociation entre pages et éléments de réponse.

la distinction entre « fait » et « vérité »

Deuxièmement, j’apprécie beaucoup ce que constat Ellul à la page 81 au sujet de l’adoption de nouvelles techniques et technologies:

le fait étant acquis, il ne saurait être question de le mettre en cause (ceci revient tout le temps ; l’intervention génétique est un fait — donc…) Il ne s’agit jamais de contester le fait à partir de la vérité. Au contraire : les découvertes biologiques étant ce qu’elles sont, quelle nouvelle éthique cela produit-il ? Tout est soumis à la force des choses et à la loi des faits.

Il s’agit ici d’une distinction faite par Ellul entre le réel (le domaine du concret, des faits, de l’évident) et le vrai (le domaine du bien et du mal, du vrai et du faux). (Il en écrit surtout dans La parole humiliée, que j’essaie aussi de lire en ce moment, mais même plus lentement…). Je trouve qu’on pourrait très bien remplacer « l’intervention génétique » par « l’intelligence artificielle » pour voir combien cette distinction pourrait nous être utile maintenant.

En fait, je dirais même que beaucoup des éléments du débat sur l’adoption de l’IA générative peut bien se résumer par cette distinction. Il y a beaucoup de monde qui dit que l’IA, c’est un fait, on a maintenant l’obligation de l’adopter dans les écoles et les universités. On n’a plus de choix, il ne reste que de décider de la manière dont on va adopter ces outils. Par contre, il y a celleux (y compris moi) qui croient que le « fait » de l’IA ne peut pas en lui-même résoudre le débat. Par contre, on a l’obligation de nous poser des questions sur la vérité : est-ce que l’IA fera du bien ou du mal dans nos salles de classe ? Est-ce qu’on devrait enseigner à nos élèves et nos étudiants de résister à l’IA dans la vie ?

J’avoue que ces questions sont compliquées, et qu’on ne peut pas nier le fait de la présence de l’IA. Pourtant, grâce à Ellul, j’ai une base théorique pour résister à l’idée que « Tout est soumis à la force des choses et à la loi des faits ».

conclusion

Il y a beaucoup plus à lire et à écrire sur un sujet comme l’IA générative. J’écris donc ce post surtout pour garder ces citations et formuler des pensées préliminaires sur ce sujet. Malgré le caractère tentatif de ce post, je peux quand-même dire avec confiance que je suis bien content d’avoir trouvé en Ellul quelq’un qui me fournit une base théorique pour mon enseignement et mes recherches.

🔗 linkblog: OpenAI and Microsoft Bankroll New A.I. Training for Teachers

Tue, 08 Jul 2025 06:41:57 -0400

Don’t know what to say here except that I don’t like any of this. Reminded of two arguments from Ellul:

First, that an effective ethics of technology considers systematic effects, not “good” uses vs. “bad” uses,
Second, that “because it exists” is not sufficient justification for adopting a technology.

Anyway, here’s the gift link.

🔗 linkblog: ‘Improved’ Grok criticizes Democrats and Hollywood’s ‘Jewish executives’

Mon, 07 Jul 2025 11:28:00 -0400

More on why we need to talk epistemology when we talk generative AI:

Musk tweaking his AI model to be more aligned with right-wing edgelords was inevitable, but there’s a broader point to be made: each AI model is a black box that supposedly gives objective answers but in reality is shaped by its owners. As more people look to AI to learn about the world, the people who control how it’s trained and how it responds will control our prevailing narratives.

🔗 linkblog: Google, de moteur de recherche à moteur de réponse

Sun, 06 Jul 2025 15:11:16 -0400

Voilà pourquoi il faut parler de la théorie de connaissance quand on parle de IA:

On est passé d’un moteur de recherche à un moteur de réponse. C’est-à-dire que les algorithmes proposent des versions rédigées à partir des données qu’ils auront collectées sur Internet, puis reformulées sans que vous ayez rendu visite aux sites contenant ces éléments de réponse à votre requête.

🔗 linkblog: Radio Télévision Suisse Pour développer l'industrie des puces, la Suisse finance la recherche plutôt que le secteur privé

Sun, 06 Jul 2025 10:29:08 -0400

🔗 linkblog: Radio Télévision Suisse Pour développer l'industrie des puces, la Suisse finance la recherche plutôt que le secteur privé https://spencergreenhalgh.com/work/2025-07-06-si-on/

🔗 linkblog: Radio Télévision Suisse Le retour de Donald Trump relance le débat sur la souveraineté numérique

Sat, 05 Jul 2025 10:35:53 -0400

🔗 linkblog: Radio Télévision Suisse Le retour de Donald Trump relance le débat sur la souveraineté numérique https://spencergreenhalgh.com/work/2025-07-05-jaimerais-parler/

🔗 linkblog: From the Backlist: Bloggers Discuss the LDS Church Compensating Social Media Influencers — Exponent II Blog

Sat, 05 Jul 2025 07:33:22 -0400

🔗 linkblog: From the Backlist: Bloggers Discuss the LDS Church Compensating Social Media Influencers — Exponent II Blog https://spencergreenhalgh.com/work/2025-07-05-hadnt-heard/

Thu, 03 Jul 2025 12:34:40 -0400

Discovered a quirky museum spotlighting local artists and found in their gift shop a zine that a former student of mine contributed to. Double win!

🔗 linkblog: Reddit turns 20, and it’s going big on AI

Sat, 28 Jun 2025 10:08:21 -0400

Reddit is a really interesting example of digital labor issues as they relate to both social media and AI. I wonder how things will go over the next few years.

🔗 linkblog: Radio Télévision Suisse A Neuchâtel aussi, les téléphones portables seront interdits à l'école obligatoire

Sat, 28 Jun 2025 09:11:52 -0400

Bon, je comprends ces soucis, mais je ne suis pas sûr que de telles interdictions soient la bonne réponse. Pourtant, vu que je suis plus ouvert à une interdiction de l’IA à l’école, il faut que je développe un peu plus ma philosophie ici.

📚 bookblog: The Sound of Mormonism: A Media History of Latter-day Saints (❤️❤️❤️🖤🖤)

Mon, 23 Jun 2025 14:13:32 -0500

I’m being a little hard on this book because it wasn’t quite was I was expecting. There’s a lot of good stuff in here despite not having some of what I was hoping to find. I do wonder if it would have been better as the original lecture it’s based on: You could hear some of the audio, and I think some of the fat could be trimmed from the manuscript. I enjoyed reading it, I just wasn’t wowed by it.

Thu, 19 Jun 2025 11:49:18 -0400

I feel like we need to talk more about epistemology—and especially the social construction of facts—in the era of chatbot LLMs.

🔗 linkblog: Teachers Are Not OK

Sat, 14 Jun 2025 09:16:54 -0400

🔗 linkblog: Teachers Are Not OK https://spencergreenhalgh.com/work/2025-06-14-bookmarked-this/

Wed, 04 Jun 2025 06:09:27 -0500

Dr. Calvin discovers the labor problems in academic publishing.

🔗 linkblog: The Internet of Consent - Anil Dash

Tue, 27 May 2025 07:38:37 -0400

Lots of good observations in here, and I need to think through the implications for digital methods research.

Sat, 24 May 2025 07:46:37 -0400

I know what this school means, and I hate feeling like I’m playing ivory tower definitions games, but it REALLY bugs me when we limit “technology” to “things with chips in them.” A wiffle ball is technology, too.

Thu, 22 May 2025 13:39:52 -0400

Who schedules an academic conference to overlap with Halloween? And how long ago did they stop having kids at home?

🔗 linkblog: Funding Cuts Are a ‘Gut Punch’ for STEM Education Researchers

Thu, 22 May 2025 06:32:16 -0400

What’s happening at the NSF is a tragedy, and I’m upset about all of these cuts. That said, I’ve long been skeptical about how the NSF has been used to promote STEM education at the expense of other worthy (but less economically productive) causes in schools. If Trump’s petty—and often cruel—cuts are a warning sign about how government can distort research priorities, there’s a deeper issue lurking in the background that we also need to wrestle with. [gift link]

🔗 linkblog: Researchers Scrape 2 Billion Discord Messages and Publish Them Online

Wed, 21 May 2025 11:18:23 -0400

I confess that I would have found this interesting in an earlier part of my career. Now, though, I’m reminded that I built that career on a methodological approach that’s uncomfortable close to surveillance, and I don’t love that.

🔗 linkblog: Duolingo CEO says AI is a better teacher than humans—but schools will exist ‘because you still need childcare’

Tue, 20 May 2025 18:30:54 -0400

🔗 linkblog: Duolingo CEO says AI is a better teacher than humans—but schools will exist ‘because you still need childcare’ https://spencergreenhalgh.com/work/2025-05-20-i-hate/

🔗 linkblog: How Miami Schools Are Leading 100,000 Students Into the A.I. Future

Mon, 19 May 2025 07:52:01 -0400

There are some critical perspectives in this piece, but certainly not enough in my book. [gift link]

Sun, 18 May 2025 08:48:03 -0400

Trying to figure out whether hate listening to the audiobook of Moneyball has implications for my suitability to teach my department’s Intro to Data Science class.

🔗 linkblog: American Schools Were Deeply Unprepared for ChatGPT, Public Records Show

Thu, 15 May 2025 12:20:56 -0400

Fascinating piece that underscores how often cheerleading voices are the only ones valued in edtech—and also how much education has been forced to respond to big tech companies simply releasing their products into the world wirhout input from those it will effect.

new publication: Jacques Ellul and educational technology

Thu, 15 May 2025 07:38:34 -0400

I’ve repeatedly referenced 20th century French technology scholar Jacques Ellul on my blog(s) since the beginning of the year. While my interest in Ellul’s work is also personal and political, I wrote back in February that one of the main reasons I’m reading a lot of Ellul right now is to add a stronger theoretical foundation to my scholarly work.

With that context in mind, I’m happy to share that my first Ellul-inspired article has just been published in the Journal of Computing and Higher Education! After I wrote this post on what Ellul had to say about the value of research, Stephanie Moore was kind enough to invite me to expand my thoughts there into a contribution for a special issue of that journal that she was putting together on “The Research We Need” in educational technology.

Turning an off-the-cuff into a medium-length journal article in a relatively short period of time was a bit of a challenge, but it was a great way to spend my end of semester, and I really appreciated the chance to pretty much immediately apply my exploration of his writing. I have found Ellul tremendously helpful for framing and voicing implicit concerns that I’ve had about research, research infrastructure, and the field of educational technology for a long time—and I think that’s what theory is good for, so this feels like a victory to me.

Anyway, there’s a read only version of the article freely available here, and here’s the abstract as a preview:

Jacques Ellul, a 20th century French academic, is best known for his writing on technique, understood as both individual means to ends and an all-encompassing system that prioritizes efficient solutions to problems. While Ellul’s terminology is not necessarily familiar in contemporary educational technology research, it represents traditional understandings of individual tools as a subset of instructional systems and captures the field’s emphasis on finding more efficient and efficacious means of teaching and learning. However, Ellul’s writing on technique is largely skeptical, creating an opportunity for educational technology researchers to turn a more critical eye towards our emphasis on solutions to problems. This essay argues that while an increased emphasis on problems over things is welcome, Ellul’s writing invites more careful consideration of what it means to solve problems. In particular, I draw on Ellul to ask three provocative questions: which problems should we solve, who should solve those problems, and is solving problems always good?

I’ve spotted a handful of typos in the manuscript that I am embarrassed to not have caught while reviewing it this past weekend (though in at least one case, I’m going to blame the manuscript proofing software, which was a real pain to work with), but despite that egg on my face, I feel pretty happy with the article. I’m also looking forward to applying Ellul’s thinking to other areas in my research.

🔗 linkblog: Oklahoma education standards say students must identify 2020 election 'discrepancies'

Wed, 14 May 2025 07:03:57 -0400

🔗 linkblog: Oklahoma education standards say students must identify 2020 election 'discrepancies' https://spencergreenhalgh.com/work/2025-05-14-ryan-walters/

🔗 linkblog: UK launching social media campaign spotlighting NIH-funded research

Sun, 11 May 2025 08:08:56 -0400

NIH funding is an important part of my employer’s budget, so I think this kind of advocacy is important. However, it rubs me the wrong way that we’re speaking up publicly about potential funding cuts and being largely silent and “well, gotta follow the law” when our marginalized students are being targeted.

Fri, 02 May 2025 06:36:36 -0400

I doubt I have time today for a longer, more nuanced post, so please enjoy this subtweet: Anyone whose takeaway from the University of Zurich Reddit study is “AI is good at convincing people!” and doesn’t even mention the HUGE ethical issues in passing is missing the point.

Thu, 01 May 2025 19:04:27 -0400

That feeling when you write an article for a lesser journal’s special issue but actually find a home in a stronger journal after the special editors reject it.

Thu, 01 May 2025 15:09:53 -0400

Today, a student from across campus explained that they used ChatGPT to identify me as a potential thesis committee member, and I am shallow enough to have had a very brief flicker of positive feeling toward the technology.

new publication: documenting a teacher group on far-right social media

Wed, 30 Apr 2025 10:01:35 -0400

I’m pleased to be able to finally share the publication in the British Journal of Educational Technology of an article that Dan Krutka and I have been working on for some time, which documents activity in a teachers’ group on a far-right social media platform (which we intentionally don’t identify within the paper). Here’s a link to a full-text, read-only version of the article, and here’s the abstract as a preview:

The affinity space framework has proven useful for explaining and understanding teacher activity on social media platforms. In this study, we explore the ‘dark side’ of teacher affinity spaces by documenting a partisan teachers’ group on an alternative social media platform. We used a mix of a priori and emergent coding to analyse screenshots of posts and comments from a public teachers’ group and group administrators’ activity on the broader platform. Findings indicate that although the group administrators began with a focus on teachers, most participants were non-teachers with political (rather than professional) concerns about US education. Furthermore, administrators both freely engaged with political talking points in their activity outside the teachers’ group and allowed the broader platform culture—including conspiratorial thinking, explicit racism and out-group villainization—to seep in. We conclude by describing how these findings correspond with the key characteristics of an affinity space, including an overlapping of affinities, a lack of concern for professional qualifications, and influence from the broader platform. These findings provide an illustrative example of how teacher affinity spaces can drift from their stated intention within the larger platform context.

I first collected this data in early 2021, so this paper has been a long time in the making. I was glad when Dan agreed to come aboard—his critical perspectives on social media were helpful here, and having someone to be accountable to helped make sure that this actually became something rather than a collection of screenshots on my computer. It took a while to find a home for this paper, which is admittedly outside of the normal realm of educational technology research, even within the “sub realm” that focuses on teacher use of social media. The route we eventually took, as indicated by the title, was to investigate this in terms of James Paul Gee’s affinity space framework and to illustrate the ways in which the characteristics of affinity spaces are not always desirable. It still took some work to get this published—we had wonderful reviewers at BJET, but the journal expressed concerns late in the process about the potentially controversial nature of our findings, and that led to some frustrating delays right at the time that we thought we’d finally made it.

Although we found evidence of explicit conspiratorial thinking, anti-Semitism, racism, and Islamophobia within the group (none of which seemingly went challenged by administrators or members), one of the more interesting findings is how banal a lot of the content was. I’m increasingly keen on using the affinity space framework to draw attention to areas of ambiguity and overlap, and I think that plays out nicely here. Within our data, we find clear evidence of far-right discourses, but there is also everyday teacher talk, and even some of the partisan discourses aren’t necessarily objectionable. I don’t want to understate the obvious red flags in the data, but I think it’s just as notable that there were cases where it was hard to draw lines between appropriate and inappropriate participation in a teacher group. The more that I study far-right-tinged groups on social media, the more that emerges as a problem in and of itself, and even though it wasn’t built for that, I appreciate the way that the affinity space framework can call scholarly attention to that sort of thing.

🔗 linkblog: Reddit Issuing 'Formal Legal Demands' Against Researchers Who Conducted Secret AI Experiment on Users

Tue, 29 Apr 2025 11:51:21 -0400

🔗 linkblog: Reddit Issuing 'Formal Legal Demands' Against Researchers Who Conducted Secret AI Experiment on Users https://spencergreenhalgh.com/work/2025-04-29-wait.-they/

🔗 linkblog: Researchers Secretly Ran a Massive, Unauthorized AI Persuasion Experiment on Reddit Users

Mon, 28 Apr 2025 12:02:15 -0400

I’ve been waiting for 404 (or someone else) to report on this so that I could rage post it (and assign it in future classes). What a terrible breach of research ethics.

Thu, 24 Apr 2025 11:01:05 -0400

A good work day is one where a colleague asks if you have a d20 in your office (I did!) because you seem like the kind of person who would (I am!).

🔗 linkblog: Who Ordered That? On AI, Education, and the Illusion of Necessity | Punya Mishra's Web

Wed, 23 Apr 2025 19:16:42 -0400

🔗 linkblog: Who Ordered That? On AI, Education, and the Illusion of Necessity | Punya Mishra's Web https://spencergreenhalgh.com/work/2025-04-23-i-would/

🔗 linkblog: ICT class gives thumbs up to new emoji submission to Unicode

Thu, 17 Apr 2025 10:12:04 -0400

This project is one of my favorite things to come out of the program I teach in, and I’m glad Meghan’s work is getting recognition!

Mon, 14 Apr 2025 07:45:59 -0400

One of the most frustrating experiences I’ve ever had with a journal appears to be coming to a happy conclusion, and that’s not a bad way to start a Monday.

🔗 linkblog: They’re putting A1 in the classrooms.

Thu, 10 Apr 2025 12:08:36 -0400

🔗 linkblog: They’re putting A1 in the classrooms. https://spencergreenhalgh.com/work/2025-04-10-this-video/

🔗 linkblog: Trump administration’s attack on university research accelerates

Thu, 10 Apr 2025 07:27:51 -0400

I don’t personally need research funding, but I work in an academic ecosystem that’s highly dependent on it. Things aren’t looking good.

Sun, 06 Apr 2025 08:48:12 -0400

I hate doing anything that seems to give Skinner any credit—especially after seeing how critical of him Ellul has been—but associating cleaning off our breakfast bar with lighting a scented candle has worked some behaviorist magic on me.

🔗 linkblog: OpenAI and Anthropic are fighting over college students with free AI

Thu, 03 Apr 2025 13:13:36 -0400

I was already planning to voice skepticism about Apple partnerships with universities in a manuscript I’m writing, but now I’ve got this to cite as well.

🔗 linkblog: Trump’s Secret Police Are Now Disappearing Students For Their Op-Eds

Thu, 27 Mar 2025 15:58:25 -0400

Masnick’s writing has never struck me as inflammatory or rushed. If he’s using this language, we should all be worried.

🔗 linkblog: Beshear vetoes bill Kentucky professors say erodes academic tenure at public universities • Kentucky Lantern

Wed, 26 Mar 2025 12:25:43 -0400

🔗 linkblog: Beshear vetoes bill Kentucky professors say erodes academic tenure at public universities • Kentucky Lantern https://spencergreenhalgh.com/work/2025-03-26-this-is/

🔗 linkblog: How the Ph.D. Project, and 45 colleges, became a target of the Trump administration

Mon, 24 Mar 2025 19:46:27 -0400

🔗 linkblog: How the Ph.D. Project, and 45 colleges, became a target of the Trump administration https://spencergreenhalgh.com/work/2025-03-24-this-is/

📚 bookblog: The Technological Society (❤️❤️❤️❤️🖤)

Mon, 24 Mar 2025 16:17:21 -0400

This is an ambitious book—probably overambitious, and I don’t agree with all of the claims, especially with 60-70 years for Ellul’s ideas to marinate in continued technological development. Yet, his ideas are valuable and prescient—I don’t buy his claims as an ontological argument, but I think they make for a compelling theoretical framework for making sense of lots of what’s happening today.

Jacques Ellul and the value of research

Fri, 21 Mar 2025 10:54:32 -0400

Last month, I wrote on both my reading up on Jacques Ellul and on concerns about how we understand the purpose and value of research. I’m continuing to read—or, rather, listen to—Ellul’s The Technological Society, and I was interested to find a passage that brought together these two ideas. Here’s Ellul, writing in the mid-twentieth century:

We have already examined the requirement of immediate applicability; here we meet it again on the state level. The state is not disinterested any more than private capitalists, but it is concerned in a different way. The state claims to represent the public interest and hence to have the duty of being a “good manager,” dispensing the public revenues only on condition that they mean something, that they pay off. Disinterested activity on the part of the state is inconceivable. Some may such that such activity should not be impossible; but in fact it is impossible. Neither individuals nor public opinion nor the structure of the state is oriented toward the acceptance of the kind of culture pure scientific research would represent.

The state demands that anything scientific enter into the line of “normal” development, not only for the stake of the public interest but also because of its will to power. We have previously noted that this will to power has found in technique an extraordinary means of expression. The state quickly comes to demand that technique keep its promises and be an effective servant of state power. Everything not of direct interest to this drive for power appears valueless. Just as financiers seek their interest in money profit, the state seeks it in power. In neither case is the motivation disinterested; technical discovery must pay off. Capitalists and state alike become impatient at delays in research, at experiments which a priori “lead to nothing,” and at the “uncertainty” of the scientist when he indulges in pure research without knowing in advance which research will pay off and which will not. Moreover, the tendency is to eliminate from the legitimate concerns of the state all sciences that have no immediate practical application: history, philosophy, grammar, and so on.

In the case of sciences susceptible of practical application, there is an immediate demand for this application. This is, of course, unfavorable to science; but it must not be imagined that it is the work of imbeciles.

The state begins by assigning a precise task to scientific research, issuing directives to the effect that it must find solutions for certain pressing problems, for example, a more rapid method to produce a part of a machine, a jet engine for aircraft, and so on. These directives are veritable commands to scientific research to summon all its resources to solve the problems as soon as possible. In a democratic system there are no sanctions against scientists who fail to fulfill the state’s demands, except suppression of financial support. A dictatorial regime, however, goes very much further to secure the compliance of the scientists. Even though it still leaves a rather broad area to personal initiative, it nevertheless tends more and more to become specific on this score.

This passage feels obviously relevant—and particularly prescient—in the context of the axes being taken to U.S. academia in this particular moment. However, what I appreciate most about Ellul’s writing here (and in other parts of the book) is the way that it gives voice to nagging concerns that I’ve had for several years about how we think of research in the United States. Even before the chaos of the current Trump administration, there’s been a tendency on the part of the U.S. system and the researchers who work within it to emphasize serving state interests at the expense of other valuable areas of academia. I’ve had to fight with reviewers and editors who want me to demonstrate the practical applications of my research findings rather than assume that there is inherent value in descriptive and exploratory work. We’ve seen the downplaying of “science that have no immediate practical application.” Funding for education research has implicitly emphasized research that in some way boosts the U.S. economy, and it’s hard for researchers with other priorities to find the same kind of support for their research.

This isn’t meant as any kind of nihilism or both-sides-ism—the past few months have clearly seen a turn for the worse. And yet, Ellul gives us a way of understanding how the “good old days” of 6 months ago were not without considerable problems either. As we bemoan the research ecosystem currently unfolding in the United States, let us have the imagination and the courage to not just long for the status quo ante but to strive for an even better system.

Thu, 20 Mar 2025 17:29:09 -0400

Look, I don’t have experience running an academic publishing company, but I don’t know if I’d activate the experimental AI summarizer on an article exploring the ethics of generative AI.

thoughts on academic labor, digital labor, intellectual property, and generative AI

Thu, 20 Mar 2025 09:59:21 -0400

Thanks to this article from The Atlantic that I saw on Bluesky, I’ve been able to confirm something that I’ve long assumed to be the case: that my creative and scholarly work is being used to train generative AI tools. More specifically, I used the searchable database embedded in the article to search for myself and find that at least eight of my articles (plus two corrections) are available in the LibGen pirate library—which means that they were almost certainly used by Meta to train their Llama LLM.

I’m not thrilled about this (though it’s hard to get too upset about something that I was confident was already happening), but as fiercely critical as I am about AI companies’ data scraping, this particular example also illustrates some of the complicated feelings and perspectives I have to sort through as I voice those criticisms. Here’s the thing: I don’t like generative AI companies using my data, but I also don’t really mind that my articles are being pirated by LibGen. I would prefer that my research be freely available to everyone, and I’m frustrated with an academic publishing system that so often locks it behind ridiculous paywalls—not so that I can profit from my own intellectual property, but so that publishing companies can.

So, my complaint here isn’t actually about piracy or intellectual property, it’s a complaint about labor. In fact, I’ve long felt that academics ought to be skeptical of AI companies’ use of our data because it so closely resembles the way that academic publishing companies use our research. I don’t mind my scholarly output being freely available to anybody—what I do mind is academic publishing companies profiting off of my work (and not giving me a penny) by locking my scholarly output behind a paywall. Sure, if my scholarly output is publicly available, I guess that means that it’s available to companies who want to use it to build their LLMs and other tools, but the core problem is still the same: Those companies are profiting off of my work (and not giving me a penny) because my scholarly output isn’t behind a paywall.

I don’t necessarily mind that Meta pirates my research publications, but I don’t think it—or anyone else!—should have to. In fact, more than one person has made the observation that Meta, OpenAI, and their ilk won’t get any more than a slap on the wrist for the same “crime” that was serious enough for the feds to drive Aaron Swartz to take his own life, and that’s a terrible world that I don’t want to live in. I don’t necessarily need to get rich off of my research publications, but if someone else is, they ought to give me a cut. In my view, the problem with generative AI tools’ exploitation of peoples’ work isn’t an intellectual property issue, it’s a deeper labor issue. That doesn’t make it any less serious, but I think the distinction here is really important.

Thu, 20 Mar 2025 08:43:23 -0400

“My anxiety is really high; taking some time off during spring break would be the healthy thing to do.” vs. “Not using spring break to get caught up is making me anxious about work I feel behind on.”

Wed, 19 Mar 2025 17:49:58 -0400

So, it turns out that spending time with Jacques Ellul’s mid-20th century critiques of systematization, abstraction, “productivity,” and “efficiency” does NOT make it any easier to write letters evaluating colleagues’ progress toward tenure.

🔗 linkblog: Etats-Unis : un chercheur français refoulé pour avoir exprimé « une opinion personnelle sur la politique menée par l’administration Trump »

Wed, 19 Mar 2025 15:15:15 -0400

🔗 linkblog: Etats-Unis : un chercheur français refoulé pour avoir exprimé « une opinion personnelle sur la politique menée par l’administration Trump » https://spencergreenhalgh.com/work/2025-03-19-cest-du/