On 9 Oct I'll be speaking at the European Journalism Centre's #NewsImpactSummit in #Warsaw. Topic of my talk: Countering #disinformation with #AI and technology. This will include presenting outcomes of @vera-ai.bsky.social.

@disinfo.eu updated its Impact-Risk Index and mapped it alongside other leading frameworks.The new study offers a comparative overview of how different methodologies define and measure #disinformation impact, highlighting similarities, gaps, and areas for improvement.

Last physical team meeting - and last group photo of a fantastic team! One month to go!

On 30 Sep and 1 Oct, we got together for the last physical consortium meeting in Thessaloniki. Our project ends in 1 month. We’re now preparing for the final steps and wrap-up. The legacy of veraAI will remain for many years to come! Thanks to a great team of diverse and committed players!

💡 ...and finally, we're sharing a concise overview of a report (elaborated for the @vera-ai.bsky.social project) mapping 𝗵𝗼𝘄 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝗳𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸𝘀 𝗮𝗽𝗽𝗿𝗼𝗮𝗰𝗵 𝗶𝗺𝗽𝗮𝗰𝘁 and where the gaps remain: www.disinfo.eu/publications...

Useful resource: the #Deepfakes Tracker Research Portal, “a rapid response project of the Social Media Lab at Toronto Metropolitan University.” Target audience, according to the providers: policymakers, researchers, journalists, the public, and other stakeholders. deepfakestracker.org

#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

Kalina Bontchava, University of Sheffield and our scientific director, rounding up and making requests / proposals for the future. Topic: what is needed and which are the primary challenges when it comes to countering disinformation.

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.

We attended #IBC in Amsterdam and showcased our work and results. Here, we report about it.

We attended #IBC in Amsterdam and showcased our work and results. Here, we report about it.

Sunday 14 Sep 2025 marked exactly 3 years of veraAI. What a journey it's been! We'll keep going though - especially as the project has been formally extended until 31 Oct 2025. For our friends @ebu.ch, however, the end is there as they are funded by another (Swiss) agency. Time for a (first) recap.

Impressions from our presence at #IBC in #Amsterdam where we have the opportunity to showcase our work and results in a booth at the @ebu.ch stand in Hall 10. Our Varvara Boboc (@digitalmethods.net) and @christosgavalas.bsky.social (ATC) are on location and sent these images.

Impressions from our presence at #IBC in #Amsterdam where we have the opportunity to showcase our work and results in a booth at the @ebu.ch stand in Hall 10. Our Varvara Boboc (@digitalmethods.net) and @christosgavalas.bsky.social (ATC) are on location and sent these images.

Our Varvara Boboc (@digitalmethods.net at University of Amsterdam) and @christosgavalas.bsky.social of ATC in action for veraAI at #IBC in Amsterdam. You can still meet them in Hall 10, stand D21.

Are you at #IBC in Amsterdam, and interested in what we have done in veraAI? If so: call in to the @ebu.ch stand (located at 10.D21 - Hall 10, space D21) where we have a presence. Come and say Hi! We'd love to see you!

Get your hands on the @ebu.ch‘s handbook on the legal and ethical obligations to developers and deployers of AI-based fact-checking tools

ICYMI: our webinar on participatory design, and how this can support in the creation of #factchecking tools - and other tools as well. All in this video 👇🏾

Alert: if you still want to participate in the Synthetic Image Detection Challenge at #MediaEval2025: the submission deadline is near: 15 September 2025 is the date! Challenge aims: push forward the research on detecting synthetic images and manipulated regions of images “in the wild”.

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.

📖 Great resource from @ebu.ch: a handbook on C2PA (Coalition for Content Provenance and Authentication) and its role in building trust and countering disinformation. 👉 Also, don't forget to check possible interlinking with @vera-ai.bsky.social tools 🔧

If you want to dive really deep into what the C2PA (Coalition for Content Provenance and Authentication) is about, and how it can support in #disinformation prevention and supporting trust, check out this handbook written by @ebu.ch colleagues. Also included: possible interlinking with veraAI tools

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.

Published today, written by @ebu.ch colleagues as part of veraAI project work: "Handbook on the legal and ethical obligations to developers and deployers of AI-based fact-checking tools". Available for reading and download right here. 👇 www.veraai.eu/posts/ebu-ha...

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.

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.

In this webinar, @nicolarighetti.bsky.social introduces the Coordinated Sharing Detection Service powered by CooRTweet (also known as CooRTweet GUI), which was developed as part of the @vera-ai.bsky.social project 👇 (Many thanks to @ebu.ch for hosting!)

Here's the recording and slides from our webinar no. 8 that took place on 3 Sep 2025, brought to you by @ebu.ch. In it, Anna Schild (@dw-innovation.mastodon.social.ap.brid.gy) and the EBU's Lalya Gaye talk about co-creation and evaluation for AI-based services for fact-checking.

Here's the recording and slides from our webinar no. 8 that took place on 3 Sep 2025, brought to you by @ebu.ch. In it, Anna Schild (@dw-innovation.mastodon.social.ap.brid.gy) and the EBU's Lalya Gaye talk about co-creation and evaluation for AI-based services for fact-checking.

We are live! Our webinar on participatory design has started. Focus: building trustworthy solutions for fact-checking and verification using AI.

30 minutes to go!

Results and lessons learned of below-mentioned work will also be presented at an @ebu.ch webinar on 3 September at 11 am CET. It's free to join but requires prior registration. So if you want to find out more about "trustworthiness by design" for fact-checking tools: this may be for you.

The @ebu.ch's Lalya Gaye and @dw-innovation.mastodon.social.ap.brid.gy's Anna Schild & Eva Lopez teamed up to develop a trustworthiness by design framework for AI-based fact-checking tools. Findings were presented at the HCI International Conference 2025 and will be published in a forthcoming paper

@ebu.ch report detailing work and approach of the C2PA (The Coalition for Content Provenance and Authenticity) and how it could support in countering disinformation.

Reminder, as it's only 1 day to go: join us on 3 Sep at 11 am CET when Anna Schild from @dw-innovation.mastodon.social.ap.brid.gy & Lalya Gaye @ebu.ch talk about the crucial involvement of end-users in designing trustworthy AI-based fact-checking services. Registration required. More 👇

As part of their veraAI work, the @ebu.ch published a report on how the C2PA (The Coalition for Content Provenance and Authenticity) works and how it could further support in countering disinformation. This includes links to our project results and approaches. More, including the whole report 👇

You can still join the MediaEval 2025 Synthetic Image Detection Challenge. Two tasks await you: A: classifying whether an image is authentic or synthetic. B: identifying and localizing regions within an image that have been manipulated. 1st deadline: 15 September 2025. Join #MediaEval25 More👇

On 3 Sep at 11 am CET the final @ebu.ch-run webinar on AI-based fact-checking tools coming out of veraAI takes place. Anna Schild from @dw-innovation.mastodon.social.ap.brid.gy & Lalya Gaye (EBU) talk about the crucial involvement of end-users in designing trustworthy AI-based fact-checking services

On 3 Sep at 11 am CET the final @ebu.ch-run webinar on AI-based fact-checking tools coming out of veraAI takes place. Anna Schild from @dw-innovation.mastodon.social.ap.brid.gy & Lalya Gaye (EBU) talk about the crucial involvement of end-users in designing trustworthy AI-based fact-checking services

You can still join the MediaEval 2025 Synthetic Image Detection Challenge. Two tasks await you: A: classifying whether an image is authentic or synthetic. B: identifying and localizing regions within an image that have been manipulated. 1st deadline: 15 September 2025. Join #MediaEval25 More👇

As part of their veraAI work, the @ebu.ch published a report on how the C2PA (The Coalition for Content Provenance and Authenticity) works and how it could further support in countering disinformation. This includes links to our project results and approaches. More, including the whole report 👇

Webinar announcement: the project is nearing its end. Hence it is time to share more results and outcomes. Join us in this webinar on 3 September in which we present ways of involving users in tool and service development. Motto: building trustworthy and intuitive AI to counter disinformation

Join us on 3 Sep at 11 am CET for the final webinar on AI-based fact-checking tools coming out of veraAI. Anna Schild (@dw-innovation.mastodon.social.ap.brid.gy) & Lalya Gaye (@ebu.ch) talk about the crucial involvement of end-users in designing trustworthy AI-based fact-checking services.

We're back after a short break! 😀

The article www.rollingstone.com/politics/pol...

Due to the summer holiday period, our activities here will be somewhat limited until mid August. Have a lovely summer everyone (assuming a) you have such a thing and b) you are located in the northern hemisphere.

Our Thomas Le Roux of Fraunhofer IDMT attended the 4th edition of the ACM International Workshop on Multimedia AI against Disinformation. It took place as part of the ACM International Conference on Multimedia Retrieval (ICMR). Here's his report about the event and a number of relevant sessions.

If you want to dig really deep and are interested in coordinated inauthentic behaviour (CIB) and network analysis: here’s a detailed report (a so-called project deliverable) on the topic, also by @fabiogiglietto.bsky.social and team. veraai-cms-files.s3.eu-central-1.amazonaws.com/D4_2_V1_0_01...

How disinformation spreads - and who is behind coordinated misinformation campaigns and sharing behaviour. One finding: the actors and notworks move on. From one topic to the next. Check out the work of @fabiogiglietto.bsky.social and team for more.

Today (w/ @ox.ac.uk @stanford @MIT @LSE) we’re sharing the results of the largest AI persuasion experiments to date: 76k participants, 19  LLMs, 707 political issues. We examine “levers” of AI persuasion: model scale, post-training, prompting, personalization, & more!  🧵: