#toread #paper The persistence of informational manipulation and the appropriation of ... by FitzGerald, Katherine M, Whelan-Shamy, Daniel, Graham, Timothy https://scholar.google.com/scholar?q=The%20persistence%20of%20informational%20manipulation%20and%20the%20appropriation%20of%20emerging%20events

I’m so happy and humbled to share that I’ve been awarded an Australian Research Council (ARC) Discovery Early Career Researcher Award (DECRA26) ($528,013 + 1 PhD scholarship funded by QUT) for my project "Democratic Resilience Online: Strengthening Public Opinion Formation Amid Digital Threats" 🥳🎉

#toread #paper Platform-independent experiments on social media by Allen, Jennifer, Tucker, Joshua A https://scholar.google.com/scholar?q=Platformindependent%20experiments%20on%20social%20media

Just found: Answer given by Executive Vice-President Virkkunen on behalf of the @ec.europa.eu from 6 March 2025 featuring our project. A fair bit has happened since then - and we are proud to say we delivered on various fronts in the fight against #disinformation

#toread #paper RIP Twitter API: A eulogy to its vast research contributions by Murtfeldt, Ryan et al. https://scholar.google.com/scholar?q=RIP%20Twitter%20API%20A%20eulogy%20to%20its%20vast%20research%20contributions

#toread #paper Multimodal narratives of climate denial: A novel, visual-fir... by Gardam, Caroline et al. https://scholar.google.com/scholar?q=Multimodal%20narratives%20of%20climate%20denial%20A%20novel%20visualfirst%20methodology%20for%20analysing%20conspiracy%20theory%20discourse%20on%20Instagram

One week to go! Still time to register! Join us for the "Meet the Future of AI" series organized by several EU co-funded projects and the @ec.europa.eu. it takes place in Brussels on 2 December. You can also participate online. Prior registration required 👇

The UW Center for an Informed Public is looking for postdocs (for 2026-2028) from across diverse disciplines whose research sheds light on the challenges of our modern information environment, promotes civic health, and/or helps people/communities navigate online spaces: apply.interfolio.com/177901

#toread #paper Untitled https://datasociety.net/library/scam-gpt/

#toread #paper Untitled https://www.sciencedirect.com/science/article/abs/pii/S2468696425000424

👋 So hello world, we did a thing. 🎁 Might be useful for some people ? #misinformation #disinformation #fakenews

#toread #paper How public involvement can improve the science of AI by Matias, J Nathan, Price, Megan https://scholar.google.com/scholar?q=How%20public%20involvement%20can%20improve%20the%20science%20of%20AI

New from me on Substack: “Source Wars and Bespoke Realities: Wikipedia, Grokipedia, and The Battle for Truth” I wrote about the high stakes power game over “reliable sources” — and the story of how Grokipedia did change its entry ab me after my Atlantic article on its hallucinations last week.

do I have any Japanese followers (or experts of the Japanese internet) who can explain why Community Notes appears to have so many Japanese super users?

#toread #paper What did Elon change? A comprehensive analysis of Grokipedia by Triedman, Harold, Mantzarlis, Alexios https://scholar.google.com/scholar?q=What%20did%20Elon%20change%20A%20comprehensive%20analysis%20of%20Grokipedia

Results. Results? Results! veraAI officially ended on 31 October 2025. But its legacy is here to stay! We make available a first set of results - available and accessible to ALL. From Keyframe Selection and Enhancement Service to the Verification Plugin. More will be added in due course.

#toread #paper Untitled https://academic.oup.com/pnasnexus/article/4/11/pgaf325/8285733

In October 2025, veraAI project members based at the University of Urbino (@giadam.bsky.social, @fabiogiglietto.bsky.social and @brupar.bsky.social) went to Brazil to present and share their research findings at a variety of events. Find out what they did in this post.

#toread #paper Emergent coordinated behaviors in networked LLM agents: Modeling the strate... by Orlando, Gian Marco et al. https://scholar.google.com/scholar?q=Emergent%20coordinated%20behaviors%20in%20networked%20LLM%20agents%20Modeling%20the%20strategic%20dynamics%20of%20information%20operations

#toread #paper What is going on? An evidence-frame framework for analyzing online rum... by Starbird, Kate, Prochaska, Stephen, Yamron, Ben https://scholar.google.com/scholar?q=What%20is%20going%20on%20An%20evidenceframe%20framework%20for%20analyzing%20online%20rumors%20about%20election%20integrity

#toread #paper Deep storytelling: Collective sensemaking and layers of meaning in U.s. elections by Prochaska, Stephen et al. https://scholar.google.com/scholar?q=Deep%20storytelling%20Collective%20sensemaking%20and%20layers%20of%20meaning%20in%20Us%20elections

#toread #paper Volume 7 issue 2 article 4 https://www.joctec.net/all-issues/volume-7/volume-7-issue-2/volume-7-issue-2-article-4

#toread #paper Large language models for data annotation: A survey https://share.google/aHRz5TZgI9RVsnDT7

🚨Special Issue Alert: Platforms, Publics, and Anti-Publics (1/9) Introduction from our guest editors Zoetanya Sujon, Harry Dyer and Felipe Bonow Soares. journals.sagepub.com/doi/10.1177/... #academicsky #socialmediasky

I really like the Better Access Framework: kgi.georgetown.edu/research-and.... Great work by an amazing team!

🚨 New preprint 🚨 Can AI agents coordinate influence campaigns without human guidance? And how does coordination arise among AI agents? In our latest research, we simulate LLM-powered AI agents acting like users on an online platform, some benign, some running an influence operation

On 31 October, our project officially ended. Now, we’re starting to wrap up. This will take until early December 2025. We‘ll continue to remain active. In particular, we will point to outcomes and results of our work. So stay tuned!

Beware Of Fake Fan Pages Spreading False Stories About Your Favorite Celebrities -- How To Spot 'Viet Spam' #prebunkingatscale leadstories.com/hoax-alert/2...

Here’s a recap of veraAI’s presence at #Disinfo2025 - one of Europe‘s primary events on #disinformation detection and analysis, organised by @disinfo.eu

#toread #paper The risks of industry influence in tech research by Bak-Coleman, Joseph et al. https://scholar.google.com/scholar?q=The%20risks%20of%20industry%20influence%20in%20tech%20research

Wow, my #AoIR2025 in Brazil was intense. Is there a word for when something drains you and energizes you simultaneously? If "exhurgizing" isn't a word yet, it needs to be.

💥 I am happy to share that my article: "Never Reveal Anything”: On Negative #Self-Branding and Relational Labor Practices of Journalists on #Platforms has been published in #Journalism Practice! 🥳 www.tandfonline.com/doi/full/10.... This articles took years - literally- to see the light of day👇

This morning at #AoIR2025 in Niterói 🇧🇷 I presented a study that’s a joint effort with @fabiogiglietto.bsky.social and colleagues. We examined how deceptive info operations evolve through cycles of trial & adaptation testing which tactics stick. Feels great to be back at @aoir.bsky.social!

This #AoIR2025 roundtable on data access is coming to an end. It started a very insightful discussion about the DSA, scraping and the future of accessing data for researchers. Thank all the speakers Josephine Lukito, Jessica Robinson, @snurb.info Richard Rogers @fabiogiglietto.bsky.social

grateful to @snurb.info for liveblogging my talk on elections, platforms & polarization at #AoIR2025🇧🇷 really nice summary of the key themes and questions discussed. read it here: snurb.info/node/3452

#toread #paper The complexity of misinformation extends beyond virus and warfare analogies by Frischlich, Lena et al. https://scholar.google.com/scholar?q=The%20complexity%20of%20misinformation%20extends%20beyond%20virus%20and%20warfare%20analogies

#toread #paper Real time detection of coordinated bots on Bluesky by Rodríguez Farrés, Pol, Masali, Athina, Cerquides, Jesus https://scholar.google.com/scholar?q=Real%20time%20detection%20of%20coordinated%20bots%20on%20Bluesky

Can large language models stand in for human participants? Many social scientists seem to think so, and are already using "silicon samples" in research. One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want. THREAD 🧵

Kalina Bontcheva of University of Sheffield presenting veraAI work at the EC's JRC conference. Here's what was developed in the project itself (in green). Other components come from previous projects InVID and WeVerify.

#toread #paper Leveraging VLLMs for visual clustering: Image-to-text mapping shows increased... by Arminio, Luigi et al. https://scholar.google.com/scholar?q=Leveraging%20VLLMs%20for%20visual%20clustering%20Imagetotext%20mapping%20shows%20increased%20semantic%20capabilities%20and%20interpretability

#toread #paper EXPRESS: A short history of misinformation-at-scale and efforts to mitigate it by Donovan, Joan https://scholar.google.com/scholar?q=EXPRESS%20A%20short%20history%20of%20misinformationatscale%20and%20efforts%20to%20mitigate%20it

A major Italian magazine thought an AI image of bullets with the alleged engravings from the Kirk assassination would be a valuable contribution to their readers. They disclosed AI use in the post, though not on the image itself, which we should expect to get fervently reused out of context.

This can be useful for Italian journalists willing to cover the "bella ciao" reference

New ADM+S submission: Research ahead of the 2025 Federal Election reveals astroturfing is mainstream in climate & energy debates. Calls for stronger laws, transparency & platform accountability to protect democracy. Read more bit.ly/41MbRYn #PoliticalAdvertising #Astroturfing #DigitalTransparency

Webinar 9 in the @ebu.ch series is out: in this final edition, the potentials of the C2PA (Coalition for Content Provenance and Authenticity) are showcased and how work undertaken can support in countering disinformation - also potentially in combination with vera.ai tools.

New handbook published by @ebu.ch - the title sums it up: "Handbook on the legal and ethical obligations to developers and deployers of AI-based fact-checking tools". Available for download.

#toread #paper Behind the screen: The use of Facebook accounts with inauthentic behavior during European elections by Oprea, Bogdan et al. https://scholar.google.com/scholar?q=Behind%20the%20screen%20The%20use%20of%20Facebook%20accounts%20with%20inauthentic%20behavior%20during%20European%20elections