Like Heather Cox Richardson’s Letters from an American, This Is Adam Kinzinger is a useful daily digest of news. I keep both in my RSS reader, where they don’t get lost among e-mails.
Monday, August 24, 2026
Profs, busted
From The Chronicle of Higher Education : “Professors Were Singled Out for Using AI in Public Writing. Now They’re Defending Themselves.” An excerpt:
Two professors recently admitted to using generative AI to assist them in writing opinion articles for major publications, igniting a debate over the standards faculty should be held to in their public writing.The justifications from Hausmann and Stankova that the article goes on to detail are, well, something. What’s the harm? And everybody’s going to be doing it anyway.
The Financial Times added an addendum to a guest essay published August 13, disclosing that its author, Ricardo Hausmann, a faculty member at the Harvard Kennedy School and director of Harvard’s Growth Lab, had used AI to “condense” his piece condemning President Trump’s tariffs, in violation of the outlet’s policy. The Financial Times is conducting an “editorial review process” to evaluate the piece in more detail, a spokesperson said in a statement. Hausmann said in an email that he hadn’t been aware of the policy when writing the article, adding that he believes AI is becoming too common for policies like that to exist.
On Tuesday, The Daily Californian reported that Zvezdelina Stankova, a teaching professor in the math department at the University of California at Berkeley, had used AI to edit an August 15 San Francisco Standard opinion article. The piece argued that the UC system was admitting underqualified students because it doesn’t consider standardized testing, and cited the fact that some students use AI to write admissions essays as a reason to bring back such requirements. A spokesperson for The Standard said in a statement that it allows AI use to assist with writing, but expects that “humans are behind every article.” The piece reflects Stankova and her colleagues’ “original analysis, research, and expertise,” the spokesperson added.
Hausmann violated the outlet’s guidelines; Stankova didn’t. But beyond a newspaper’s code of conduct, what ethical responsibility — if any — do professors have when it comes to AI in their public-facing work, given that students are generally expected to refrain from using it to write?
Related reading
All OCA AI posts (Pinboard)
[I’m not interested in focusing OCA on AI in academia. But I couldn’t pass up this item in the news.]
“That’s very perspicacious of you”
Today’s Dustin : Kevin reads a thesaurus. See also these two — 1, 2 — Zippy strips.
Related posts
Beware of the saurus : Rewording : Rogeting
Sunday, August 23, 2026
Mutts vibrations
[Mutts, August 23, 2026. Click for a larger view.]
Today’s Mutts takes up a great American song. I’m glad I didn’t miss the title (aka “throwaway”) panel.
Related reading
All OCA Beach Boys posts and Mutts posts (Pinboard)
[I know: “Good Vibrations” did not appear on the Pet Sounds album. For anyone who needs it: the album cover.]
Writing and AI at Harvard
At The Atlantic (gift link), Tyler Austin Harper writes about “Why the Closing of Harvard’s Writing Center Matters.” He suspects that Harvard’s decision will give other schools permission to close their writing centers:
The decisions that Harvard makes create a permission structure for other institutions to follow suit — after all, if one of the most venerable schools in the country doesn’t need a writing center in order to be “elite,” why would a public university, or a regional liberal-arts college, or a community college need one?I see it differently: I think that emphasizing the presence of a writing center might offer a lower-tier school a smart and snarky way to present itself to prospective students. On a campus tour, for instance: “This building also houses our writing center. As you may know, Harvard University recently got rid of its writing center. Unlike Harvard, we still have a writing center, because we believe that students benefit from working on their writing with another human being, not with a chatbot.”
As Harper points out, Harvard is all in on AI. A development that just made the news, postdating his essay: Harvard is now selling a course on entrepreneurship “taught” by AI replicas of business faculty (The New York Times , gift link). Only $699!
The prose of Harvard’s unrelated Teach with Generative AI page (AI-generated prose?) does not inspire confidence. For instance:
Harvard has consistently tried to embrace new technology across our classrooms, residential and virtual. GenAI has been no different.It’d be Harvard that’s been no different, not AI.
Behold, a fused sentence:
Faculty and students have access to a range of tools. Some of these tools are free and open to all faculty, students and staff behind Harvard Key (Harvard Sandbox) other tools require a license or approval.My favorite:
Out of this there have been several learnings worth considering.And I’m sure there’ll be many more.
Another Yankee Diner
[2494 Flatbush Avenue, Marine Park, Brooklyn, c. 1939–1941. From the NYC Municipal Archives Collections. Click for a much larger view.]
Did you know that there’s a Brooklyn neighborhood called Marine Park? I didn’t. Did you know that at one time there were at least Yankee Diners in Brooklyn, this one, the one above, and a third whose tax photograph seems to be unavailable? I did.
The nameless dining room to the left looks (emphasis looks) as if it was part of the Yankee enterprise. And notice the bakery/restaurant to the right. And those cars. There’s more food on this block, coming soon.
Related posts
More photographs from the NYC Municipal Archives (Pinboard)
Saturday, August 22, 2026
Tell me you’ve never read my blog without telling me that you’ve never read my blog
In today’s mail:
We recently came across Orange Crate Art — a quietly artful personal blog.And so on.
That's exactly the kind of artful personal blog we built [redacted] for — an AI content team that keeps creating for you. Tell it what to make — newsletters, videos, podcasts, or articles — and how often; it researches fresh sources, picks what matters, and produces editable content on schedule, so morning musings draft with the coffee.
I replied with a link to my Pinboard list of OCA posts tagged AI (three of those posts are on the front page right now) and three more words: “No, no, no.”
[“Morning musings”? And they “draft with the coffee”? Seriously?]
Recently updated
Things I learned on my summer vacation: Today’s Zippy prompted me to seek out a photograph of a Guggenheim Museum bathroom that I linked to in 2010. That photograph has disappeared from the Internets, but I found another.
Today’s Saturday Stumper
I found today’s Newsday Saturday Stumper, by Ben Zimmer, wildly difficult. Daily puzzles — Newsday, the Los Angeles Times, the syndicated New York Times — seem to me increasingly easy. I can often do an entire puzzle, or nearly an entire puzzle, from Across clues alone. But the Saturday Stumper seems to get more and more difficult and unfun. Humph.
Some clue-and-answer pairs of note:
1-A, five letters, “Little-used.” This clue starts the puzzle off with a bang: three or four different letters could plausibly fill the answer’s first square.
1-D, eight letters, “Social media ‘Spare me your replies.’” Never heard of the answer. I must need to partake of more social media. Oh, wait — I don’t.
2-D, fifteen letters, “Signals profound change.” Surprisingly easy to see.
4-D, four letters, “Barcelona bros.” Yes, but. Pretty arcane, imho. 1-D!
8-D, fifteen letters, “Unforgettable, in a way.” I liked this one.
10-A, six letters, “Papers’ attention-getter.” I thought it had to be a slang term for a giant headline.
19-A, nine letters, “Weird Al parodied it with ‘Bologna’ (1979).” Good grief.
25-A, six letters, “What the first jeans were, per its etymology.” I was sure the answer had to do with Nîmes and denim , but nothing would fit, not even after I took a seamripper to the grid.
26-D, letters, “Wellington museum models.” Can you tell that I found this puzzle exasperating?
40-A, seven letters, “Slide-show subjects.” Cute. The clue, not the subjects.
42-A, six letters, “Legendary rock singers.” Rock on.
44-A, four letters, “Name on a Scotch bottle above a clipper ship.” I know it, but there is no reason anyone in 2026 should be expected to know it.
My favorite in this puzzle:
30-A, eight letters, “Seller of Tova cosmetics to QVC (2002).” Because it’s utterly bizarre and patently ridiculous — and because it was my starting point in the puzzle.
No spoilers; the answers are in the comments.
Subscribe to:
Posts (Atom)
.jpg){kind=link}