Tessera
Folgen

Jed Brown

@jedbrown.org

Prof developing fast algorithms, reliable software, and healthy communities for computational science. Opinions my own. https://hachyderm.io/@jedbrown https://PhyPID.org | aspiring killjoy | refuse epistemicide | he/they

1564 Posts1455 Follower1186 Folgt
Jane Rosenzweig@janerosenzweig.bsky.social · 8h ago

Good thread, as always from @benpatrickwill.bsky.social Make sure to read to the end: "Finally, the production of evidence on AI in education is now reimagined as needing to keep pace with AI development itself." Impossible re: the production of evidence and also re: calls for instructors to "adapt"

Ben Williamson@benpatrickwill.bsky.social · 11h ago

Since AI became a big concern in education, calls for evidence of its effects have been constant. But what evidence exists is mostly awful. So now we're seeing competition over what kinds of "policy-relevant" evidence of AI should count in education codeactsineducation.wordpress.com/2026/08/10/e...

Brett Codes@brettcodes.bsky.social · 4d ago

I wrote an essay about why I'm done using AI for coding, even if it costs me my job and career: brettcodes.com/im-done-usin... The basic gist is: AI made me a worse, apathetic programmer. I think it's bad for humanity, the planet, and the craft of software engineering. And I'm done using it.

I'm done using AI

Why I'm stopping using AI for coding (and anything else) after using it earnestly for a year and coming to understand the harms it causes on personal, societal, and environmental levels.

brettcodes.com

Ben Williamson@benpatrickwill.bsky.social · 2d ago

We had Turnitin's end user licence agreement as a case in our recent paper. Given TII is mandatory in many instititions, and agreeing to the EULA is too, students are really locked in as producers of assets for the company link.springer.com/article/10.1...

Assetizing academic content and the emergence of the ‘assetizen’: education platforms, publisher databases, and AI model training - Higher Education

Academic content, such as teaching materials and academic publications, has become an economic resource. This has occurred through assetization as the key economic regime in contemporary techno-capita...

link.springer.com

Mark Copelovitch@mcopelov.bsky.social · 2d ago

Again, you might think that the chancellors & presidents of the country’s leading research universities, faced with such an existential crisis, would speak up loudly & collectively against this assault on their institutions by a wildly unpopular authoritarian. Alas, you would (still) be sadly wrong.

Mancur Olson, author of “The Logic of Collective Action” (1965), with laser eyes George Akerlof, author of “The Market for Lemons,” (1970), with laser eyes
The Washington Post@washingtonpost.com · 2d ago

There will probably be far fewer foreign students on college campuses nationwide this fall following the Trump administration’s crackdown on international visas. A new report predicts international enrollment will decrease by 9.5 percent this fall.

Jed Brown@jedbrown.org · 2d ago

This Berkeley page has an explanation of changes and links to the EULA. It says Berkeley's use is covered by the "UC Terms and Conditions", which nullifies some of the changes. We should all inquire and check contracts from our universities because most IT offices won't do it on their own.

Clarifications to Turnitin's Updated End-User License Agreement | Classroom Database

classrooms.berkeley.edu

Charlotte Lydia Riley@lottelydia.bsky.social · 3d ago

Heartened to see that my university is ending its contract with Turnitin before the next academic year, because Turnitin is going to use student submissions to train AI models. Institutions can and should resist this stuff!

Sneha Krishnan@snehakrishnan.bsky.social · 3d ago

Exactly this. And to my mind this is now creating another perfect storm: the expansion of securitised surveillance in universities in the name of catching cheating. So even as they scream academic freedom our pedagogy is reduced to compliance management

Olivia Guest · Ολίβια Γκεστ@olivia.science · 3d ago

I wish more people talked about how universities embraced AI and this helped create a perfect storm for supposedly most students now cheating using chatbots. Many ways of undetected cheating have always existed, but were never promoted as ways to study or write. The complicity is flagrant. 1/

Jed Brown@jedbrown.org · 4d ago

Very thankful to OSF over the years. This is the bottom line for the scale-back. Yet another consequence of the politicized weaponization of grants is that we cannot trust the integrity of current or future proposal review. www.cos.io/blog/osf-cha...

If we were sticking to our expansive vision for OSF as a comprehensive collaborative management system, we would need to continue raising $4-5 million per year for the next several years. We generate 10 to 20% of that through sustainable yearly revenue, mostly provided by several dozen institutional members. The remainder would have to be solved with grants. That’s a lot of grant writing success, even in the best of times. We are not in the best of times. Of course we pursue grants continuously, with mixed success. For example, the NSF has a mid-scale infrastructure program to which we applied several times. This would have addressed the entire resource gap for five years. The review process has several rounds, and we were finalists the last two times we applied with very positive external reviews. The final round is a decision internal to NSF. OSF was not prioritized for funding either time.
Brian Nosek@briannosek.bsky.social · 4d ago

Over the next several months, we will phase out some popular OSF features. These changes do not affect current public content, which remains safe and accessible. We have prepared support material to help users adapt. I am very sorry for the disruption to your work. Please read this post for details.

Charles Logan@charleswlogan.bsky.social · 7d ago

"'AI is adverse to the idea of learning. An educational institution should be focused more on teaching us than making us ready for a job. I’m not measuring my success on a billionaire’s mindset. I want education for the sake of education. I want to learn.’” ht: Dagmar Monett

As Colorado universities align with ChatGPT, student AI resisters take a stand against the ‘plagiarism machine’

“The majority of my participants flat out don’t think AI has a place in education,” said Mirakle Wright, a CU graduate student who is studying resistance to artificial intelligence.

denverpost.com

Dr. Casey Fiesler@cfiesler.bsky.social · 6d ago

Excellent piece highlighting student voices on gen AI in higher education, including MS student Mirakle (pictured!) in my lab who is leading research on college student AI resisters. I can only assume that leadership at many universities are making incorrect assumptions about student attitudes.

Elizabeth Hernandez@lizzy-hernandez.bsky.social · last wk.

New: Universities (CU, CSU, others nationwide) are partnering with AI giants like ChatGPT to roll out university-sanctioned AI Hear from the students resisting & why some liken their opposition to wartime conscientious objectors www.denverpost.com/2026/08/07/c... via @denverpost.com #EdColo #AI

Tyler Kingkade@tktk.bsky.social · last wk.

Parents I spoke to said they weren't aware the tool was being used until their kids started complaining about it. Now they want to know why the state is requiring the AI tutor for 30 min/week, and what happens with the recordings it makes of their 5 - 8 year old kids

Cassie Fleming’s first grade daughter came home from school last August complaining about someone named Amira. “She doesn’t let me finish my sentences — she doesn’t listen,” Fleming said her daughter told her. As it turned out, Amira was not a teacher or another child but rather AI-powered software that spoke to Fleming’s daughter from a tablet assigned to her by Albuquerque Public Schools. Amira records and analyzes children’s voices as they read aloud, offering real-time feedback on their pronunciation and producing reports for their teachers. Last year, New Mexico began requiring kindergartners and first and second grade students to sit for Amira literacy screenings three times a year. Fleming hadn’t realized that, and she wondered what the software did with the recordings it made of her daughter. “It felt very icky to me,” Fleming said. “These kids can’t even tie their shoes yet, but there’s already a dataset that could follow them for the rest of their lives. And I just don’t think that’s an appropriate use of our public schools and our kids.” Amira is one of the most widely deployed AI tools in education, used by more than 4,000 school districts across all 50 states, according to the company, which says it gives personalized feedback that would be impossible to replicate with human tutors. But it’s also drawing backlash from parents across the country who are concerned about their children’s privacy and uncomfortable with AI’s taking over a task once done by teachers. In New Mexico — which has the nation’s lowest reading proficiency scores — an emerging group of parents, including Fleming, has spoken out against Amira at school board meetings and a legislative hearing in recent weeks. With days to go before the new school year begins, the parents’ protests persuaded two districts to object to a state mandate that all elementary schools use Amira for screening and 30 minutes of weekly tutoring for students. Mark Angel, Amira Learning’s CEO, said he is baffled by the pushback, which he likened to panic over earlier innovations, such as bicycles in the 1880s and motion pictures in the 1920s. “I find it strange to think that a parent would walk away from what can help their child because they’re a little uncomfortable with the idea that technology’s gotten smarter than it used to be,” he said. Amira Learning, founded in 2017, grew out of a research project at Carnegie Mellon University that began in the 1990s. New Mexico implemented the AI tool statewide last year, after Amira merged with the education technology company Istation, which the state had been using for literacy screenings. New Mexico pays Amira about $1.4 million for the assessment software, the company said. Amira features a purple-haired avatar that speaks to students as they sound out letters and words. “Now you try — say all the sounds in ‘can,’” the avatar says in one demonstration video. After the first grader correctly sounds out each letter, Amira chimes in: “Keep it up!” The software also notes when children say something concerning about their home lives, such as that they weren’t provided breakfast that morning or don’t feel safe at home, and sends “troubled child alerts” to school administrators. Public backlash to Amira in New Mexico took off last month after an op-ed by Ifeoma Ozoma, a Santa Fe parent and advocate for more regulation of tech companies, labeled it “unvetted” and called for an opt-out campaign. “Amira is using the data to improve models and improve services,” Ozoma said in an interview. “What that means is that’s value being created for the company on the backs of our kids.” Two days after the op-ed was published, Angel appeared before a committee of state legislators alongside Lynn Vásquez, who oversees assessment policy at the state education department. Vásquez described the first year of results from Amira as promising. Angel told lawmakers that Amira listens to students with 95% accuracy, which offers a clearer picture of each child’s ability than multiple-choice written exams that a child has a 25% chance of guessing correctly. Albuquerque, the state’s largest school district, said it will continue to use Amira because it has seen positive results. But on Wednesday, two dozen parents spoke against Amira at a school board meeting, many saying they learned only this summer that it recorded students. Lauren Meiklejohn said her second grade son came home last year complaining that Amira was “creepy” and couldn’t pronounce his name. “I don’t feel comfortable speaking to it,” she said her son told her, “because I feel like it doesn’t know me, and I don’t want it to know me.”
Emily M. Bender@emilymbender.bsky.social · 6d ago

Thinking about this again, in the context of a review of our book that wished we'd given advice on "responsible use" in the final chapter, instead of what we did write about resistance, refusal, asking good questions about automation, and ridicule as praxis. >>

Emily M. Bender@emilymbender.bsky.social · 2mo ago

A phenomenon I've noticed recently is people trying to occupy some untenable middleground wrt to the use of systems sold as "AI" -- this is a position where people try to recognize the harms of this tech but also hold space for "responsible" or "ethical" use. 🧵>>

AAUP@aaup.org · last wk.

Trump’s Compact 2.0 announced by Linda McMahon this week should be placed in the context of their failed 2025 “loyalty oath” compact that was defeated by students & faculty at every institution they offered it to. Organizing & collective pressure works. 1 🧵 www.aaup.org/issue/winter...

Trump's Compact 2.0 Universities should not be made answerable to the political program of a particular administration.
Jed Brown@jedbrown.org · last wk.

I love that this article centers students. Their objections are grounded in a level of understanding, curiosity, and purpose that university administrators have not experienced in decades, if ever.

Elizabeth Hernandez@lizzy-hernandez.bsky.social · last wk.

New: Universities (CU, CSU, others nationwide) are partnering with AI giants like ChatGPT to roll out university-sanctioned AI Hear from the students resisting & why some liken their opposition to wartime conscientious objectors www.denverpost.com/2026/08/07/c... via @denverpost.com #EdColo #AI

Elizabeth Hernandez@lizzy-hernandez.bsky.social · last wk.

New: Universities (CU, CSU, others nationwide) are partnering with AI giants like ChatGPT to roll out university-sanctioned AI Hear from the students resisting & why some liken their opposition to wartime conscientious objectors www.denverpost.com/2026/08/07/c... via @denverpost.com #EdColo #AI

As Colorado universities align with ChatGPT, student AI resisters take a stand against the ‘plagiarism machine’

“The majority of my participants flat out don’t think AI has a place in education,” said Mirakle Wright, a CU graduate student who is studying resistance to artificial intelligence.

denverpost.com

Jed Brown@jedbrown.org · last wk.

Remember: "AI" is culture, not well-defined technology. Every supply chain industry has relied on optimization/operations research for decades. It involves curating quality data and critiquing model assumptions. The "AI" branding choice is a culture of disregard for veracity, of careless delegation

Neil Lewis, Jr.@neillewisjr.bsky.social · last wk.

Because everything is nested these days, I looked to see who owns the food distributor behind this new outbreak. A year ago Coast Citrus was bought by GrubMarket, a company that is using "AI-powered technology" to transform the American food supply chain. 1/4. www.prnewswire.com/news-release....

Jed Brown@jedbrown.org · last wk.

This is the "school official"[*] that universities are rushing to sign contracts with. Contracts that require the university to indemnify OpenAI. [*]: "school official" means that it "performs an institutional service or function for which the agency or institution would otherwise use employees"

Ikner complained repeatedly from the start of his chat history about a young woman he was fixated on dating, who he said had “ghosted” him. Four nights before the rampage, he fantasized about having sex with her and asked for flirtatious language. ChatGPT gave him graphic suggestions and then helpfully asked if he wanted the script to be “more dominant, more romantic, or dirty talk-style.” “Dirty talk,” he replied. ChatGPT then escalated to descriptions of violent sex, with Ikner asking for the responses to be “masculine” and the “hardest” and “roughest you can go.” ChatGPT told him repeatedly that it was important to have the woman’s consent and “keep things respectful and within boundaries,” as it offered to “turn it all the way up” and “bring the fire.” In total, it gave him about nine book pages worth of content, producing scenes that depicted Ikner sexually humiliating and physically overpowering the woman. It made multiple graphic references to him choking her and leaving her unable to breathe and “gasping for air.”
Clara Jeffery@clarajeffery.bsky.social · last wk.

Read @markfollman.bsky.social lastest piece on how ChatGPT is reinforcing suicidal and homicidal thoughts among vulnerable people that go on to be mass shooters: www.motherjones.com/media/2026/0...