dietram a. scheufele

The fact that #AI is doing more and more of the researching, writing, and reading of fiction and non-fiction books for us might spell serious trouble for the medium … at least for scholarly work. nytimes.com/2026/03/19/boo…

"#AI requires no ramp-ups, no meetings, and absolutely no emotional support ... I feel somewhat embarrassed to admit how tempting this is."

The first openings for the fourth cohort of #civicscience fellows were just posted: civicsciencefellows.org/stories/2026-2… #scicomm #scipol #engagement

"@Waymo adds 4 more cities to its robotaxi service, now 10 total (@Tesla : still 0)." electrek.co/2026/02/24/way…

People form beliefs not only as individual agents, but as members of social groups. A new paper finds that children (4-6 years old) who belonged to a group were more convinced by evidence that supported their ingroup’s belief (and were less convinced by evidence that opposed their ingroup): nature.com/articles/s4146… This is exciting evidence that the Identity-based Model of Belief extends to young kid and novel groups! All kinds of identities shape our beliefs about the world. sciencedirect.com/science/articl…

"Many of these bills provided agencies with supplemental appropriations, some of which could be directed toward research and development." Nimble universities that are open to change fare better in modern funding ecosystems: ncses.nsf.gov/pubs/nsf26310 #highered #scipol

As TikTok settles one case in LA, never forget that it has harmed millions of kids in many ways; addiction is just one. And they know it. And we know they know it because we have so many damning quotes from their employees. Read them here: afterbabel.com/p/industrial-s…

AlphaGenome is our latest & most advanced genomics model published in @Nature today including making the model & weights available to academic researchers. Can’t wait to see what the research community will do with it. Congrats to the team on our newest front cover! #AI4Science

"[L]arge language models (LLMs) demonstrate strong reasoning capabilities in closed-world settings, they struggle in open-ended and dynamic environments." #AI #information #scicomm #scipol arxiv.org/abs/2601.12538

Scientists who use AI have published three times more papers, received five times more citations, and reach leadership roles faster than their AI-free peers. But science as a whole is paying the price, the study suggests. Not only is AI-driven work prone to circling the same crowded problems, but it also leads to a less interconnected scientific literature, with fewer studies engaging with and building on one another. It's a classic social dilemma: what's good for individuals is bad for the collective. science.org/content/articl…

Okay so, we just found that over 50 papers published at @Neurips 2025 have AI hallucinations I don't think people realize how bad the slop is right now It's not just that researchers from @GoogleDeepMind, @Meta, @MIT, @Cambridge_Uni are using AI - they allowed LLMs to generate hallucinations in their papers and didn't notice at all. It's insane that these made it through peer review👇

"[I]ndustry ties [among authors, reviewers, and editors] are associated with a topical focus away from impacts of platform-scale features." Our evidence base for effects of #socialmedia is confounded by undisclosed #COIs. MUST-read by Bak-Coleman et al. arxiv.org/pdf/2601.11507

“There’s always opportunity,” @ArizonaState's Michael Crow said in an interview on redesigning "a flawed system" of #highered. “And there’s opportunity now.” Important and provocative read ... regardless of one's partisan preferences. wsj.com/us-news/educat… #scipol #scicomm

Don’t be distracted by large N size: It’s difficult to reach any conclusion based on a single survey item — esp. on topic as multidimensional as attitudes/beliefs about emerging tech / AI. See Pew multi-item trends increasing reservations across years + strong proportion DK.

"Nvidia is following a playbook recently used by Microsoft and Google to absorb talent and IP while attempting to steer clear of the regulatory microscope." Responsible #AI needs transparency and cooperation, but tech has zero incentive to play nice. theinformation.com/articles/nvidi…

“If we want a robust society, we need more robust dialogue, and that must include the right to insult or to offend.” Not new, but more timely than ever … youtu.be/xUezfuy8Qpc?si…

A little too close to home to be funny ...

There are the "big" awards that pad CVs and annual reports, and then there are recognitions like this that remind us that teaching is a privilege ... in spite of the current state of @UWMadison and U.S. #highered. #instructors" target="_blank" rel="nofollow noopener">housing.wisc.edu/undergraduate/… #teaching #scicomm #LSC251

Our 2017 @theNASEM report on “Human Genome Editing: Science, Ethics, and Governance" (nationalacademies.org/our-work/human…) was designed to prevent exactly this kind of rogue science. One of the many reasons why we need industry-independent regulatory frameworks … wsj.com/tech/biotech/g…

"Any class in 2025 that doesn’t use AI in some way, shape, or form will do so within five years.” Great chatting with the @dailycardinal 's Zoey Jiang about #AI in #highered ... dailycardinal.com/article/2025/1…