Vi holder på å skrive et letter of concern til tidsskriftet nå. Som forresten er samme svenske tidsskrift som i denne saken ⬇️ Hovedforfatteren av den finske studien er en av to internasjonale samarbeidspartnere i den ulovlige forskningen på OUS.

Skremmende tråd Storbritannia er nå kollektivt så langt nede i kaninhullet at det er grunn til å sitere Pratchett: “Do you think it's possible for an entire nation to be insane?” ― Terry Pratchett, Monstrous Regiment At Navratilova ikke har fått fyken fra BBC etter i praksis ha karakterisert 1/

Har sittet og grått på SAS-flyet til Stavanger fordi jeg leste novellen "Story of your life" i samlingen The story of your life and others av Ted Chiang... OG DET ER KANSKJE DEN VAKRESTE NOVELLEN JEG NOEN GANG HAR LEST?!?!?!😭😭😭

[Blogged] AI academic search needs better frameworks for understanding and evaluation. These three librarian projects are a start open.substack.com/pub/aarontay...

A group of Ecuadorian fishermen survived a US drone strike, were detained at gunpoint, hooded, phones wiped, ship blown up, disappeared to El Salvador, then released without charge. “They knew we were fishermen. Even the Salvadorian authorities said things had been handled very badly.”

#overhørt student med selvinnsikt: "jeg er så dårlig på å bruke KI så jeg nekter å bruke det"

Selv deltar jeg på seminar om skeive data 2. Juni nettopp for å sikre god forskningsdatahåndtering og legge godt grunnlag for skeiv forskning ved OsloMet. Men det er nå bare meg da.

The wholesale failure of the mainstream so-called "AI ethics" project is because it is instead a successful process of creating industry stooges, who walk amongst us, hollowing out future colleagues and posturing provocatively about the inevitability of the future theintercept.com/2019/12/20/m...

Yeah, this obsession with language is a trick to stymie so-called nonnative speakers. Make them feel less confident and waste money on proof readers or other non-issue solutions. If your English is not understandable your colleagues should help anyway and it would have been caught years before.

/competence porn\ …that’s a new one, and it makes so much sense.

The Pitt is 'competence porn' in more ways than one aftermath.site/the-pitt-hbo-c...

Very disjointed to read “wearing them seems like a violation of the social contract” next to a breakdown of price and battery life.

Dette blir spennende! selveste Maia Kobabe kommer til Stavanger i september! Passer veldig godt når jeg skal undervise emnet mitt, Global Sexual Politics igjen!

US tech giants pressured the EU into making environmental data about the impact of individual data centres confidential - so journalists, activists and local politicians CANNOT know the actual environmental toll of the local data centres confidential. www.techpolicy.press/how-big-tech...

Its such a testament to the lightning in a bottle of Heated Rivalry that it did this but somehow that made it MORE dramatic and potent.

"The field of cultural game studies holds a strong position at Tampere University, with three professors, two university lecturers and several doctoral researchers working in the field. Together, they form a vibrant research community that attracts Finnish and international visitors to the campus."

Synnøve Vereide Trampe blander seg inn i den såkalte transdebatten, og Aftenposten gir henne en helside. Folk som ikke selv er trans er i en helt utrolig grad selvoppnevnte eksperter på våre liv, og driver frisk cherry picking av selektive rapporter. Rapporter som passer med deres fordommer. 1/

Jeg skrek HØYT ute i gata i dag da jeg så nyheten om Genderqueer til Stavanger. 🏳️‍⚧️ Og bare TUSEN TAKK til KRF som veldig raskt viser hvorfor det her er viktig. Jeg blir nesten rørt. Skulle trodd Kapittel hadde booka han for å drive markedsføring 😍

Absolutt alle alarmklokker bør ringe når politikere prøver å bestemme hvilke bøker bibliotekarer skal få lov til å ha i bibliotekene sine. Spesielt når disse politikerne faktisk er farlig nærme å ta makten ved neste stortingsvalg.

So between the 2 camps information literacy peeps supporting undergrad and evidence synthesis librarians, it feels to me there is a third camp missing. Librarians who care about tools that give you reasonable recall and precision for narrative reviews. I belong to this camp...(13)

Jippi! 6. mai kommer min kollega Siri Fürst Skogmo fra UINN til @oslomet.bsky.social for å snakke om prosjektet "Verden på norsk", en åpen digital fagressurs hvor man kan utforske oversettelse av skjønnlitteratur og hvordan ideologier spres. #digitalhumaniora #bibliosky

In the end, I am making the same point for a profession that claims to be expert at search, there seems to be surprising lack of interest in testing or valuing retrieval. There's a small specialisation - evidence synthesis that focuses on a niche use case that isnt relevant for most people (12)

Ironically on the opposite end of spectrum to you get librarians involved in evidence synthesis. For them they are too strict cos they need 100% recall & find ai search not good enough. Reality is most researchers lie between 2 exteme of needing 100% recall (ES) & freshman need for precision@10 (8)

Not into the AI-generated podcast summaries of #chi2026 paper sessions – that authors *cannot* opt out of. But was thinking that perhaps this could make papers more accessible to the general public. Turns out the AI-podcasts are only available to ACM premium members... wtf...

ACM #chi2026 authors: ACM is doing slopcasts for your papers. Just because they don't offer an opt-out doesn't mean you shouldn't protest this surrender to wasteful mediocrity. Feel free to repurpose what I wrote about the slop 'summaries' codeberg.org/mdingemanse/... >

it probably is a ddos attack, but done by their own slop code it's a pretty common thing that ai generated code generates 100x api calls instead of 1, seen lots of cases of vibecoders doing this and running up saas costs; one of their devs admitted it in a post mortem for an outage 10 days ago

in which bsky admits the outage 10 days ago *was* their fucking code after all pckt.blog/b/jcalabro/a... archive.is/AKfKP

I love how The Pitt has subtly but effectively said what needs to be said about the health insurance crisis, anti-vax madness and "free birth" insanity. All while celebrating science. Bravo.

This is what going on the internet used to feel like

We have seen this before, with the kidnapping of Maduro or the mass protests against ICE raids – fake images of real events. While these synthetic images fictionalize reality, they are not necessarily meant as disinformation. Rather, they both illustrate and cater to wide-spread expectations 1/

In a great short essay, @robhorning.bsky.social has some sharp observations on the same phenomenon: «AI slop clips often latch on to news stories and try to show viewers what the news would be like if it could be stripped of its context and presented as maximally engaging to an individual user.»

Hvem kom på at det var en god idé å invitere nettopp Nicolai Tangen til å åpne et kulturelt arrangement...?

Our team received a report of intermittent app outages at about 11:40pm PDT on April 15, 2026. They worked through the night to mitigate a sophisticated Distributed Denial-of-Service (DDoS) attack, which intensified throughout the day.

«Hegseth published a book in 2020, American Crusade, which considered the destruction of Muslim holy sites to reclaim them for Christianity. He has tattoos re the Christian crusades. Hegseth has incorporated religion into his work at the Pentagon, hosting monthly evangelical services »

In Pulp Fiction, the Bible verse is explicitly called out as asking for a peaceful shepherd who avoids violence and refuses to believe the world should be ruled by the strong. The film is about Jules realizing that he doesn’t understand the verse he’s quoting until he gives up violence.

Secretary of Defense Pete Hegseth quoted a violent and fictional Bible passage nearly word-for-word from Quentin Tarantino’s Pulp Fiction in his Wednesday night worship service at the Pentagon.

[Read] Which AI Tools Work Best for Research? Using Librarian and Student Perspectives to Inform a Rating Rubric tandfonline.com/doi/full/10.... used a "rubric-based approach to assessing AI tools across three key areas" Undermind.ai & the paid Elicit emerged as top performers. " (1)

As an aside if a tool is keyword matching based and your tool is using "semantic search" (vector embeddings) you shouldn't be surprised overlap is less and users perceived the later results as more "serendipitous". (6)

Benytter muligheten til å anbefale alle som har lyst til å jobbe i bibliotek, med samlinger eller annen kulturvirksomhet å søke denne masteren her. Du blir også ram på å analysere teknologi. Ja, den er også samlingsbasert og kan tas ved siden av jobb, så her har man masse gode muligheter:

In other words, AI resentment maps pretty directly to how precarious your job situation is. For Gen Z, this is the worst entry level job market in decades; no wonder they hate AI, which they are routinely told is the reason why there's no work. The rich, meanwhile, are just fine with more automation

AI is profoundly unpopular. A recent NBC News poll found that among 18-34 year-olds, AI's net favorability rating is -44. *Negative 44*. Those are basically serial killer numbers. It's not much better among women 18-49. (Men over 50 and upper class are the only ones who like AI, and just barely.)

Last week, a 20 year-old man threw a molotov cocktail at Sam Altman's mansion; two days later, people fired a gun at it. Earlier that week, someone fired gunshots into an city councilman's house who approved a data center. Why the AI backlash has turned violent:

Mediefolk som får tyn på sosiale medier – feks *enkelte* kjent fra dax18 og Debatten – er raske til å avfeie kritikk som oppgulp fra en liten kohort med idioter Pussig nok er dette aldri et alternativ når en liten kohort med idioter mener noe ikke-mediarelatert som er *veldig* lett å avfeie

Kanskje det er dumt å la en facebookgruppe på størrelse med en gruppe for kjøp og salg av utstyr til hest diktere offentlig samtale og mediebildet i to uker i strekk selv om algoritmene i sosiale medier gjør sitt ytterste for å løfte frem de som er sinte og høylytte

Har en venn som kaller alle som insisterer på å tre KI nedover hodene på ansatte for New Age Management. OG JEG ELSKER DET.

WaPo website replaced search with AI and it's going great

Huh www.reuters.com/legal/govern...

Vet du hva som er enda flauere enn å ha en oljefondsjef som prøver å leke influenser?   Å ha en oljefondsjef som enten er for tjukk i huet – eller for arrogant – til å følge helt basic spilleregler absolutt alle influensere har fortstått i over et tiår allerede.

Microsoft starts removing "unnecessary" Copilot buttons from Windows 11 apps. www.theverge.com/news/909640/...