Does engaging with the arts help to slow the aging process? According to this new paper from Fancourt et al the answer is yes! Alas, the paper is actually unsuitable for such claims. Let’s examine by discussing how you'd establish whether arts engagement actually caused future health! 1/18 🧵

absolutely wild to see people out here saying that it is responsibility of peer reviewers to check for hallucinated references. it is hard enough to persuade people to read articles and provide reviews. expecting them to spend hours cross reffing literature is unreal.

Disappointed that #KingsSpeech said nothing about SLAPPs, which enable the powerful to stymy activists & stop journalists publishing stories in the public interest. The legal sector has successfully lobbied to sink efforts to legislate, for now. Excellent piece in the @newstatesman1913.bsky.social

Here it is. But this analysis does not prove the issue is with social sciences, if anything it makes the case stronger the issues is with SSRN.

NEW: A serious staffing shortage — of the Trump admin's own making — is delaying the agency's ability to send billions to universities around the country, leaving labs reeling in the meantime. “I thought we were at rock bottom”, a senior NIH official said. “We are below rock bottom now.”

Strong policy from arXiv! Hallucinated citation = ban for a year from arXiv, and 'subsequent arXiv submissions must first be accepted at a reputable peer-reviewed venue'. [repost: I incorrectly characterized Dietterich before - he is the vice-president of arXiv's external advisory council].

Kinda wild that arXiv is announcing on social media first this 1-year-ban for authors who use LLM-slop (including hallucinated-references). But presumably they'll put something up official soon. Right, @tdietterich.bsky.social ?

We are implementing a similar policy at @arxiv.bsky.social. If there is incontrovertible evidence of LLM slop in a paper, this means the authors did not take the time to read the LLM output and we can't trust anything else in the paper. Penalty is 1 year ban from arXiv followed by...

A potential caveat: SSRN hosts two entities associated with life science journals, one associated with the Lancet and the other with Cell. I believe a decent chunk of their content is from those sources (maybe 20%?). I wouldn't frame/understand this effect as a disciplinary difference.

SSRN, the social-sciences preprint site, has five times higher hallucinated/fake citations than arXiv, bioRxiv etc, according to a new study (not yet peer-reviewed). By @stokel.bsky.social | @nature.com www.nature.com/articles/d41...

The local elections in one chart. Labour has been squeezed between the Greens in young wards and Reform in older working-class wards www.economist.com/britain/2026...

After reading this, the AI bioweapons threat that seems most likely to me, is that some state actor or other well-resourced malevolence, will use AI to help design new kinds of toxins [which they might be trying to make anyway]. www.nature.com/articles/d41...

AI assistance is becoming more expensive for researchers, making them think twice about using it go.nature.com/3PbuQbH

This survey is interesting tho notably mingles answers from professional physicists and magazine readers. Answers can differ by background. Eg the astrophysics/cosmology/gravity researchers are far less keen on MOND than the respondents who are 'in another field' or merely 'science enthusiasts'.

Virologist accused of starting COVID-19 will fight U.S. ban on funding | Science | AAAS www.science.org/content/arti...

This preprint on hallucinated citations suggests a greater problem on SSRN (reaching 1.9% of total by August 2025) than arXiv, bioRxiv, or PMC (variously 0.2-0.4%). Intriguing if methods hold up; study restricts itself to looking for references with non-existent titles. arxiv.org/abs/2605.07723

Lawsuit harks back to fights over text-mining. Scientists wanted to use software to mine data from research articles at scale but publishers often insisted they had to buy extra licenses for this. 'The right to read is the right to mine'! was the researcher chant. www.nature.com/articles/d41...

I think it's worth people knowing that Enfield has a Labour MP and a Conservative councillor who are *both* called Bambos Charalambous.

Best evidence yet that LLMs can match human capabilities in doing science.

"I work at the US Army Medical Research Institute of Infectious Diseases, which has been working on hantavirus vaccines since around the 1980s." www.nature.com/articles/d41...

it took me too long to realize that this was "horseshit" censored and not new slang i didn't know about yet where "horses" meant bad

On fears that attention spans are shrinking. Yes, we do flit from task to task more quickly and that does worsen performance. But our underlying capacity to pay attention seems undiminished. Ditch distractions and we can recover focus. By David Adam | @nature.com www.nature.com/articles/d41...

Are attention spans really shrinking? asks @nature.com It feels wrong to summarise this, but: A: ppl do flit from task to task more often than in the past but "there is little evidence that the brain’s fundamental ability to concentrate has been impaired" www.nature.com/articles/d41...

Well, well. A science publisher (Elsevier) decides to sue an AI firm over copyright infringement. Lawsuit with other publishers (Macmillan, Hachette etc.) vs Meta for training Llama models on copyrighted content, in Common Crawl and then from Sci-Hub among others. www.reuters.com/sustainabili...

How much of the scientific literature is generated by AI? www.nature.com/articles/d41...

“You can rebuild a building. But if attacks on universities become a normal thing, then they can happen in any future stupid war. And this idea is much more destructive than attacking a single building." — Ali Gorji

Cool cell biology for a Friday: ribosomes that run with proteins made from 19, rather than 20, amino acids. "a feat akin to rewriting one act of a Shakespearean play without a common letter like ‘R’ while keeping the text intelligible" - Elie Dolgin in @nature.com www.nature.com/articles/d41...

Can octopuses, squid and cuttlefish reveal fundamental principles of brain evolution and function? More and more neuroscientists think they can and are flocking to research these enduringly fascinating creatures. My latest for @nature.com www.nature.com/articles/d41...

Craig Venter (who has died aged 79) redrew the boundaries of biology — sequencing DNA at unprecedented speed, engineering synthetic life and charting ocean microbes. by @ecallaway.bsky.social | @nature.com www.nature.com/articles/d41...

Worrying reading www.bbc.co.uk/news/article...

Red blood cells modified with ‘click’ chemistry can snap together to form clots that staunch bleeding in seconds. Tested so far in rats, this produces clots stronger than both natural clots & a commercial product used to stop bleeding. @nature.com www.nature.com/articles/d41...

The Trump administration is downsizing US science by historic margins — but it's not just via funding or workforce cuts. Our new @nature.com analysis reveals the government has cut more than 100 scientific advisory panels across major agencies. w/ @alexwitze.bsky.social @dangaristo.bsky.social

Yeah when machine learning only has to do one specific task well it is perfectly possible to make it via ethical means.

Have grants become more competitive lately? 🧪 At least some of the uptick in submissions (that we've reported on @nature.com) seems driven by AI-assisted proposal writing -- funders, like journals, are flooded Great comment piece by @jameswilsdon.bsky.social & co: www.nature.com/articles/d41...

My story on the NSB terminations:

A first-author slot on an article sold by a paper mill costs a median value of nearly US$800. Price range from $57 to > $5,600. @miryamnaddaf.bsky.social in @nature.com on @reeserichardson.bsky.social, @abalkina.bsky.social analysis of thousands of papermill ads www.nature.com/articles/d41...

Don't know what happened to this, but the website is down, and the DOI record has been wiped from Zenodo, so I guess making a public complaint works.

Should you worry about AI killing us all? I looked at the evidence so you don't have to 🧪 www.nature.com/articles/d41... @nature.com Ft. @garymarcus.bsky.social @sashamtl.bsky.social @gleave.me @andreamiotti.bsky.social @anthonyaguirre.bsky.social @ghadfield.bsky.social among others!

When @karaswisher.bsky.social promotes health misinformation while attacking it, we have a problem. www.the-geyser.com/kara-swisher...

A lot of Orbán's American fans are saying that this election proved he was never an authoritarian in the first place. This is completely wrong — and, in fact, betrays a complete misunderstanding of both Hungarian politics and modern authoritarianism. Here's why.

Scepticism about the hype that ‘ghost murmur’ quantum sensors can spot heartbeats from miles away. www.science.org/content/arti...

Another fake disease AI thought was real was "Kyloren syndrome." It showed up in Google's AI Overview. It was from a hoax paper that @neuroskeptic.bsky.social wrote to expose a predatory journal. See "The garbage'll do", July 2017, in my hoax anthology, Stinging the Predators: bit.ly/StingPred

🚨 For nearly two years - and even now, depending on the day and model - AI models could tell you all about bixonimania, a totally made-up disease in a paper funded by “the Professor Sideshow Bob Foundation". My @nature.com story on AI's fallibility in medicine www.nature.com/articles/d41...

Many of the claims made regarding the enhanced efficiency of AI tools are misleading; this one isn't. Large language models (LLMs) are *very* efficient at spreading false information. #ArtificialIntelligence #LLMs 🧪 www.nature.com/articles/d41...

I am biased but this is the greatest and possibly only Artemis II video you need 🌝 www.youtube.com/shorts/Tc2vN...

My colleague & I teamed up w/ Grounded AI to look for fake/hallucinated references. Our analysis suggests that at least tens of thousands of publications from 2025 likely contain such references. More detz at @nature.com, w/ input from @gcabanac.cpesr.fr, @mhmdhsini.bsky.social & @kowb.bsky.social.

Until you change the incentive structure of academia you will continue to see people fear losing their status and use AI as a way to fast-track productivity bc academia doesn't currently value the meaningful production of knowledge, only the performance of producing it.

This is the inevitable consequence of there being far too many scientists and journals whose only goal is to publish as many papers as possible. Before these papers would have been junk, now they're AI generated junk.

Tens of thousands of papers published in 2025 contain fake/hallucinated citations, an analysis by Nature and Grounded AI suggests By @miryamnaddaf.bsky.social and Beth Quill | @nature.com www.nature.com/articles/d41...

The question about fake references should NOT be “What can be done?” We KNOW what can be done: human peer reviewers and editors check the damn references individually, by hand. The question should be, “Why are so many publishers NOT doing those basic checks?” www.nature.com/articles/d41...