Lightnews — Scholar-powered news

Reposted by Camilla Holm Soelseth

Aaron Tay @aarontay.bsky.social · 1h

Found this on mailing list. Still hard to visualize what's going on

Camilla Holm Soelseth @supercamilla.bsky.social · 3h

De betaler ikke, men TikTok gir penger til kontoer som lager innhold hvis de er registrert som en viss type konto og produserer innhold med viss terskel for metrics over tid.

1 1

Camilla Holm Soelseth @supercamilla.bsky.social · 6h

Perfect! Thank you and @matti-lamela.bsky.social for this text! This will be read (and most probably) used in a future article I am working on 🎉💚

2

Reposted by Camilla Holm Soelseth

Amanda Wasielewski @amandawasielewski.com · 1d

The 30th anniversary anthology PDF is now published online:

Read "What is Digital Humanities and what's it Doing in the Department of ALM (Archive, Library, Museum)? by myself and Matti La Mela here: uu.diva-portal.org/smash/get/di...

uu.diva-portal.org

1 2 6

Reposted by Camilla Holm Soelseth

Tim Dickinson @timdickinson.bsky.social · 10h

The Portland ICE protest camp literally has a rack full of inflatable costumes for demonstrators who want to try one on...

68 880 4.3K

Reposted by Camilla Holm Soelseth

Jathan Sadowski @jathansadowski.com · 12d

What an unsurprising decision by corporations that desperately need to make money with AI. If every other company with a chatbot isn't already doing this, then expect them to be following Meta's lead soon. The chatbot is not your friend; it's a corporate listening device. www.ft.com/content/22f7...

Please use the sharing tools found via the share button at the top or side of articles. Copying articles to share with others is a breach of FT.com T&Cs and Copyright Policy. Email licensing@ft.com to buy additional rights. Subscribers may share up to 10 or 20 articles per month using the gift article service. More information can be found at https://www.ft.com/tour.
https://www.ft.com/content/22f7afc3-8ac0-4ca1-9877-fd3f8ddcc986?desktop=true&segmentId=7c8f09b9-9b61-4fbb-9430-9208a9e233c8#myft:notification:daily-email:content

Meta will use conversations people have with its chatbots to personalise advertising and content across its platforms, in a sign of how tech companies plan to make money from artificial intelligence.

The owner of Facebook, Instagram and WhatsApp on Wednesday said it would use the content of chats with its Meta AI to create advertising recommendations across its suite of apps.

“People will already expect that their Meta AI interactions are being used for these personalisation purposes,” said Christy Harris, privacy and data policy manager at Meta.

9 160 380

Reposted by Camilla Holm Soelseth

bleary @bleary.off-the-records.com · 19h

If anyone needs me I will be in the museum, lying down next to the bog bodies.

Did people really memorize phone numbers before cell phones, or is that just a movie thing?
2? Questions
I was watching some old shows from the 90s and noticed people would just dial numbers from memory - like they'd call their friends or family without looking anything up.
Made me wonder if that was actually normal back then? Did people genuinely have all their important numbers memorized, or did most folks keep a little address book or written list nearby?

1.4K 4.5K 22K

Reposted by Camilla Holm Soelseth

John Scalzi @scalzi.com · 8h

I note some opposition to this proposition, ie "But surely [author] holds up better than every other!" and "there is no perceptible variation in literary opinion based on disparate life experiences," and in both cases: oh, honey. Okay.

John Scalzi @scalzi.com · 12h

The answer to "how does [author] hold up?" is always "for whom?" Bester holds up fine to me, but I'm a 56-year-old white dude who has a vested interest in the genre. My take on his hold-uppedness will be different than, say, a 15-year-old BIPOC girl approaching science fiction for the first time.

Publius Maximus @publiusmaximus.bsky.social · 12h

How does Alfred Bester hold up?

16 27 400

Reposted by Camilla Holm Soelseth

Alice @starinthetitle.bsky.social · 7h

You actually don’t have to read books if you don’t enjoy it, reading is a hobby for fun, it isn’t inherently a morally superior way to absorb information/stories and you are not automatically a better person for doing it 🤷‍♀️

Rachel Feder @rachelfeder.bsky.social · 1d

Tell me your most unhinged literary opinion, as a little treat

1 5 37

Reposted by Camilla Holm Soelseth

Micah @rincewind.run · 18h

Discworld QOTD, from Monstrous Regiment

“Keep out of the way of officers, ‘cos they ain’t healthy. That’s what you learn in the army. The enemy dun’t really want to fight you, ‘cos the enemy is mostly blokes like you who want to go home with all their bits still on. But officers’ll get you killed.”

Micah @rincewind.run · 1d

Discworld QOTD, from Hogfather

The late (or at least severely delayed) Bergholt Stuttley Johnson was generally recognized as the worst inventor in the world, yet in a very specialized sense. Merely bad inventors made things that failed to operate. He wasn’t among these small fry. Any fool could make something that did absolutely nothing when you pressed the button. He scorned such fumble-fingered amateurs. Everything he built worked. It just didn’t do what it said on the box. If you wanted a small ground-to-air missile, you asked Johnson to design an ornamental fountain.

6 96 530

Reposted by Camilla Holm Soelseth

wendy copium @alicebennett.bsky.social · 16h

one thing LLMs have revealed is that the belief that there is a magical incantation that will give you what you want is not just confined to pick-up artists and freemen on the land

Des Fitzgerald @desfitzgerald.bsky.social · 18h

Seeing people nonchalantly post stuff like makes me feel like I'm going crazy honestly.

A post from an academic on blue sky that says: "I'm learning about "prompt engineering" when asking LLMs to extract data. I now add this to all instructions:

Style
Be concise, analytic, and specific (cite page/figure if provided). If essential info is missing, keep going but flag Unclear and exact data needed. Never invent data.

#academicsky"

12 160 570

Reposted by Camilla Holm Soelseth

Fagatha Ghostie @joelnb.bsky.social · 1d

Don't make me tap the Robert Pattinson quote, Leto.

“I always say about people who do method acting, you only ever see people do the method when they’re playing an assholes,” Pattinson said in an interiew with Variety in November. “You never see someone being lovely to everyone while they’re really deep in character.”

26 680 3.1K

Reposted by Camilla Holm Soelseth

Joe Bak-Coleman @jbakcoleman.bsky.social · 7h

Open science rhetoric to justify jamming agi into predatory journals and convert rando data to “papers”

Samuel Moore @samuelmoore.org · 19h

"Tasks that once took months of manual work — from curating datasets and checking compliance to creating metadata and publishable outputs — are now completed in minutes by the AI Data Steward"

Another case of commercial publishers looking to replace library labour with their junk AI.

90% of Science Is Lost: Frontiers’ revolutionary AI-powered service transforms data sharing to deliver breakthroughs faster

Frontiers, the open-science publisher, is tackling this problem with the launch of Frontiers FAIR² Data Management, the world’s first all-in-one, AI-powered ser

www.frontiersin.org

3 7 16

Camilla Holm Soelseth @supercamilla.bsky.social · 6h

Relatable

Amanda Wasielewski @amandawasielewski.com · 6h

When I’m in this critical AI/AI ethics bubble it feels like there are a ton of ppl in academia thinking critically about AI but then I go and sit in a seminar full of ppl in the natural sciences and realize how many academics are uncritically excited to have LLMs do their work

10

Reposted by Camilla Holm Soelseth

Samuel Moore @samuelmoore.org · 19h

"Tasks that once took months of manual work — from curating datasets and checking compliance to creating metadata and publishable outputs — are now completed in minutes by the AI Data Steward"

Another case of commercial publishers looking to replace library labour with their junk AI.

90% of Science Is Lost: Frontiers’ revolutionary AI-powered service transforms data sharing to deliver breakthroughs faster

Frontiers, the open-science publisher, is tackling this problem with the launch of Frontiers FAIR² Data Management, the world’s first all-in-one, AI-powered ser

www.frontiersin.org

3 8 29

Reposted by Camilla Holm Soelseth

Thorsten Möhlmann @thorstn.bsky.social · 8h

«The input does not cause the output in an authorial sense, much like input to a library search engine does not cause relevant articles and books to be written (Guest, 2025). The respective authors wrote those, not the search query!» via @olivia.science via#2 @irisvanrooij.bsky.social - thank you

Olivia Guest · Ολίβια Γκεστ @olivia.science · 4d

important on LLMs for academics:

1️⃣ LLMs are usefully seen as lossy content-addressable systems

2️⃣ we can't automatically detect plagiarism

3️⃣ LLMs automate plagiarism & paper mills

4️⃣ we must protect literature from pollution

5️⃣ LLM use is a CoI

6️⃣ prompts do not cause output in authorial sense

5 Ghostwriter in the Machine
A unique selling point of these systems is conversing and writing in a human-like way. This is imminently understandable, although wrong-headed, when one realises these are systems that
essentially function as lossy2
content-addressable memory: when
input is given, the output generated by the model is text that
stochastically matches the input text. The reason text at the output looks novel is because by design the AI product performs
an automated version of what is known as mosaic or patchwork
plagiarism (Baždarić, 2013) — due to the nature of input masking and next token prediction, the output essentially uses similar words in similar orders to what it has been exposed to. This
makes the automated flagging of plagiarism unlikely, which is
also true when students or colleagues perform this type of copypaste and then thesaurus trick, and true when so-called AI plagiarism detectors falsely claim to detect AI-produced text (Edwards, 2023a). This aspect of LLM-based AI products can be
seen as an automation of plagiarism and especially of the research paper mill (Guest, 2025; Guest, Suarez, et al., 2025; van
Rooij, 2022): the “churn[ing] out [of] fake or poor-quality journal papers” (Sanderson, 2024; Committee on Publication Ethics,

Either way, even if
the courts decide in the favour of companies, we should not allow
these companies with vested interests to write our papers (Fisher
et al., 2025), or to filter what we include in our papers. Because
it is not the case that we only operate based on legal precedents,
but also on our own ethical values and scientific integrity codes
(ALLEA, 2023; KNAW et al., 2018), and we have a direct duty to
protect, as with previous crises and in general, the literature from
pollution. In other words, the same issues as in previous sections
play out here, where essentially now every paper produced using
chatbot output must declare a conflict of interest, since the output text can be biased in subtle or direct ways by the company
who owns the bot (see Table 2).
Seen in the right light — AI products understood as contentaddressable systems — we see that framing the user, the academic
in this case, as the creator of the bot’s output is misplaced. The
input does not cause the output in an authorial sense, much like
input to a library search engine does not cause relevant articles
and books to be written (Guest, 2025). The respective authors
wrote those, not the search query!

14 34

Reposted by Camilla Holm Soelseth

Arnfinn Pettersen @arnfinnp.bsky.social · 21h

Sjælveste Magnus Lunde. IYKYK. Og hvis ikke, kan du prise deg lykkelig.

UiO-ansatt anklages for å spre propaganda

På Facebook forsvarer han Russlands invasjon av Ukraina. På UiO er han ansatt for å lage en øvingsplattform for studenter. Den skal lære dem hvordan Norge beskytter seg mot dataangrep. Svært proble...

www.universitas.no

3 8 42

Reposted by Camilla Holm Soelseth

Social Media Lab @socialmedialab.ca · 15h

Better late than never ... "News outlets broadly reject Pentagon rules before signing deadline. The Washington Post joined the New York Times, Newsmax and CNN in refusing to sign the restrictive new policy." www.washingtonpost.com/business/202...

News outlets broadly reject Pentagon rules before signing deadline

The Washington Post and other major news outlets have refused to sign the Defense Department’s restrictive new media policy, citing First Amendment concerns.

www.washingtonpost.com

2 7 37

Camilla Holm Soelseth @supercamilla.bsky.social · 7h

Det virker som utrolig mange tror det er ekte - og det fører til interaksjoner. Så jeg har lurt på det same. Men må lese meg mer opp på hvordan tjene penger via TikTok for å kunne gjøre et anslag her

1 1

Camilla Holm Soelseth @supercamilla.bsky.social · 16h

Å ffs. Gi meg tilbake Eurovision 😭

ESC Discord @escdiscord.com · 22h

🚨 The vote to determine Israel's participation in #Eurovision 2026 will no longer take place in November, the EBU has confirmed today.

Instead, an "open and in-person discussion" will take place at the EBU General Assembly in December. No vote is currently planned.

5

Camilla Holm Soelseth @supercamilla.bsky.social · 18h

"No. Wo... Women are... are supposed to just put up with this shit, but no.

Oh, no, we have to kill them.

...

We cannot keep killing the men who wrong us!"

- Ursula, Bad Sisters

3

Camilla Holm Soelseth @supercamilla.bsky.social · 21h

Den mest bananas kontoen på TikTok nå er den som er laget med Sora og storyline er at en gravid dame prøver å bestige Mount Everest men så må hun føde og babyen dør og hun må begrave ungen på Mount Everest. Og dette fortelles gjennom 100 videoer og de er alle KI-genererte og det er så whæck.

2 16

Reposted by Camilla Holm Soelseth

Aaron Tay @aarontay.bsky.social · 21h

As of 2025 academic "ai discovery" is essentially RAG over localised central indexes. The question is whether the existing central index eg Exlibris CDI, EDS will secure agreement from content providers to allow RAG over. Already Elsevier + others have opted out of Summon/Primo RA (2)

1 2 4

Camilla Holm Soelseth @supercamilla.bsky.social · 21h

Jeg tror låta "Opalite" treffer meg så godt fordi det er akkurat sånn det er å jobbe i IT-avdelinga på OsloMet, og jeg får kanskje ikke forska så mye for øyeblikket men helt greit når det nesten alltid er kake i lunsjen og memes i kanalene.

Vil gjerne anbefale denne arbeidsplassen og gi den 10/10

1 7

Reposted by Camilla Holm Soelseth

Laura Bassett @lebassett.bsky.social · 1d

Roses are red
Fossils are old
www.newsweek.com/jordan-peter...

Jordan Peterson diagnosed with chronic condition after exposure to mold

Peterson's daughter said her father has a "genetic predisposition that causes the immune system to have an inability to identify and detoxify mold."

www.newsweek.com

36 32 240