Verdraagzaam Breukelen 😅

Enjoy 💚! #fomo

Het antigif tegen tech-halleluja is een onafhankelijke en sterke journalistiek

Spijkers & koppen:
"Het antigif tegen tech-halleluja is een onafhankelijke en sterke journalistiek" @felienne.bsky.social
www.volkskrant.nl/columns-opin...

Reality Re-Imag(in)ed. Mapping Publics’ Perceptions and Evaluations of AI-Generated Images in News Contexts The rapid advancement of generative AI technology has transformed the creation and dissemination of visual content in journalism, raising critical questions about its impact on news production, cre...

We deden er onderzoek naar. (Wel binnen nieuwscontext): www.tandfonline.com/doi/abs/10.1...

Hoe een journalist uitglijdt op door AI verzonnen citaten Gebruik van AI-programma’s als ChatGPT als schrijfassistent leidt regelmatig tot missers. Na incidenten in media, wetenschap en advocatuur, ging nu ook journalist Peter Vandermeersch in de fout. „Ik h...

4. On the journalistic sector as a whole, where trust is already under pressure and such mistakes further fuel doubts about control, care, and transparency.

Full article, in Dutch:
www.nrc.nl/nieuws/2026/...

Reporting by @mennovdbos.bsky.social

2. On the misquoted experts, whose names become linked to words they never said – and that continue to circulate via LLMs.
3. On Mediahuis, the organization where he played a key role for many years.

I was able to contribute to this investigation myself, but I believe this kind of use has repercussions on at least four levels:

1. On Peter Vandermeersch himself, still a role model for many in journalism.

In 15 of the 53 blog posts he wrote, there are quotes that cannot be found in the publications from which Vandermeersch claims to have taken them. He admits his mistake, but to me this case shows how AI is sometimes used 'blindly,' and that irresponsible use does indeed have consequences.

Hoe een journalist uitglijdt op door AI verzonnen citaten Gebruik van AI-programma’s als ChatGPT als schrijfassistent leidt regelmatig tot missers. Na incidenten in media, wetenschap en advocatuur, ging nu ook journalist Peter Vandermeersch in de fout. „Ik h...

Interview in NRC on bad AI use in journalism[In Dutch]
Ex-editor-in-chief, Peter Vandermeersch is currently “Fellow Journalism & Society” @ Mediahuis. In that role, he published dozens of fictitious quotes attributed to real experts. Mediahuis has temporarily suspended him www.nrc.nl/nieuws/2026/...

In 15 van de 53 blogposts die hij schreef, staan citaten die onvindbaar zijn in de publicaties waar Vandermeersch ze vandaan zegt te hebben, zoals nieuwsartikelen en wetenschappelijke onderzoeken. Vandermeersch geeft zijn fout toe, maar de case toont hoe 'blindelings' AI soms gebruikt wordt.

Hoe een journalist uitglijdt op door AI verzonnen citaten Gebruik van AI-programma’s als ChatGPT als schrijfassistent leidt regelmatig tot missers. Na incidenten in media, wetenschap en advocatuur, ging nu ook journalist Peter Vandermeersch in de fout. „Ik h...

Oud-hoofdredacteur van o.m. NRC, Peter Vandermeersch, is tegenwoordig 'Fellow' bij Mediahuis. Daarbij publiceerde hij tientallen fictieve citaten van echte deskundigen. Mediahuis heeft hem tijdelijk geschorst. Onderzoek door @mennovdbos.bsky.social 👇
www.nrc.nl/nieuws/2026/...

Comprehensive overview by @stuartathompson.bsky.social & @ajcardia.bsky.social 👏

Cascade of A.I. Fakes About War With Iran Causes Chaos Online The technology has been used to create misleading fakes before. But never at this scale.

This might just be the tip of the iceberg…

The New York Times identified over 110 unique A.I.-generated images and videos from the past two weeks about the war in the Middle East. 👇🏼

www.nytimes.com/interactive/...

@imec-mict-ugent.bsky.social

Hoe snelle opmars van AI ook in Vlaanderen tot ongemak leidt: gaan we er wel slim mee om? | VRT NWS Nieuws Amper een paar jaar na de introductie van AI-tools als ChatGPT, Gemini of Copilot is nu ruim 4 op de 10 Vlamingen gewoontegebruiker. Dat is ongezien, zo blijkt uit het jaarlijkse imec.digimeter-onderz...

Nieuwe #digimeter is uit.
Voornaamste bevinding: 4 op de 10 Vlamingen is 'gewoontegebruiker' geworden van AI-tools als ChatGPT en Copilot. Ook de bezorgdheid over AI groeit. Zo vreest meer dan de helft van de jonge gebruikers door AI minder zelf na te denken dan vroeger.👇
www.vrt.be/vrtnws/nl/20...

How AI Is Turbocharging the War in Iran Intelligence, targeting and damage assessments are accelerating thanks to military versions of software now remaking business and daily life.

Disappointed but not surprised.

-> How AI is turbocharging the war in Iran: 🫠
"AI tools are helping gather intelligence, pick targets, plan bombing missions and assess battle damage (...) AI helps commanders choose the best weapon for each objective." www.wsj.com/tech/ai/how-...

The #QuitGPT campaign is picking up serious momentum - especially with news of OpenAI’s deal with the US war department.

I wonder: Quitting ChatGPT and cancelling subs is one step. But what do we do when OpenAI models are baked into so many existing tools and other digital infrastructures?

Also cultural differences and institutional/university differences: some uni's really push LLM use or institutionalise it, whereas other inst/unis discourage LLM use - doesn't make the whole responsible usage discussion easier... not for aca. journals and unis (no silver bullets etc.)

Indd - in focus groups, native speakers were stating that they are not using LLMs in their work, and they expect their co-authors not to use them either. At the same time, non-native s were using LLMs quite often. We also saw regional differences, but indd largely reproducing exist. power structures

🔥 New preprint 🔥
Guidance over guidelines? Unpacking the uses and concerns of generative AI in Communication Science 👇
osf.io/preprints/so...

@ascor.bsky.social @aimediademlab.bsky.social @ddc-sdu.bsky.social @theoaraujo.bsky.social @damiantrilling.net

Link to preprint:
osf.io/preprints/so...

Five Takeaways From the ‘First-of-its-kind’ International Survey About the Use of Generative AI in… If Generative AI seemed inevitable to you one day and off-limits the next, you are not alone. In this preprint, results of a…

💥 𝐍𝐞𝐰 𝐩𝐫𝐞𝐩𝐫𝐢𝐧𝐭 𝐨𝐧 𝐆𝐞𝐧𝐀𝐈 𝐮𝐬𝐞 𝐢𝐧 𝐂𝐨𝐦𝐦 𝐒𝐜𝐢𝐞𝐧𝐜𝐞 💥

International survey of 1138 scholars across 77 countries - plus four focus groups - maps the what, how and why. 69.2% already use it for at least one research task, especially writing and literature review.

5 takeaways:

medium.com/@hannes.cool...

"Silence is the loudest sign of a regime in crisis."

Iran's internet blackout has now passed the 24-hour mark with national connectivity flatlining at 1% of ordinary levels. 👇

🧐 Interesting to see the launch of SPUR (Standards for Publisher Usage Rights), a new coalition from BBC, Sky News, the FT, the Guardian and The Telegraph.

They’re building shared, global standards so AI can access high-quality journalism responsibly:
LINK: www.bbc.co.uk/mediacentre/...

Inside the $265mn war over who writes America’s AI rules Silicon Valley rivals in spending blitz for control of Congress and state capitols

"Groups campaigning for and against more AI regulation are amassing at least $265mn in collective financial firepower, according to FT analysis. This is $100mn more than the main parties’ fundraising vehicles for House elections had raised by this time in 2022."
www.ft.com/content/c182...

De Correspondent en Follow the Money gaan samenwerken om onafhankelijker te worden van Amerikaanse big tech Al jaren schrijven we kritisch over onze afhankelijkheid van Amerikaanse techbedrijven. En wij vinden: wie daarin geloofwaardig wil zijn, moet ook bereid zijn het zelf anders te doen. Daarom starten w...

Vandaag kondigen De Correspondent en Follow the Money aan dat ze gaan samenwerken om onafhankelijker te worden van Amerikaanse big tech ✨
@casperwolfert.bsky.social
decorrespondent.nl/16752/de-cor...

Washington Post Begins Laying Off More Than 300 Journalists

Democracy dies when laying off journalists...
www.nytimes.com/2026/02/04/b...

🇪🇺🌐 The future of social media may be decided in Europe A survey of European AT Protocol projects, including 4 mini case-studies

"The future of social media may be decided in Europe", @laurenshof.online writes. He argues that developers have shown that it is possible to build independent platforms on AT Protocol, yet suitable funding mechanisms are largely absent.👇
#digitalsovereignty
newpublic.substack.com/p/the-future...

Academic Freedom Under Attack This panel discussion addresses the growing global assault on academic freedom—a cornerstone of democratic societies now under increasing threat from authoritarian regimes.

🎙️ Our lab is organizing a panel discussion with @natalihelberger.bsky.social and special guest @oscarwestlund.bsky.social on the global challenges posed to academic freedom. Join to learn why defending it matters!

📍 SPUI25 (and online)
📅 24.02 | 17:00
Sign up for free: www.spui25.nl/programma/ac...