Technology and AI to Counter Disinformation

By Zlatina Marinova on May 19th, 2021 in Events, News

An Online Workshop

16 June 10:00 CEST – 14:00 CEST

WeVerify Technology Demonstration

17 June 10:00 CEST – 13:00 CEST

The WeVerify project is honoured to invite and welcome you to a free two-day workshop on 16 and 17 June 2021. Day one, 16 June, focuses on the role of technology and tools in the verification and fact-checking process. It will approach the topic from various angles and positions, namely the side of facilitators such as the European Commission and specific projects and initiatives in the field; practitioners using tools and technologies for verification and fact-checking on a regular basis; and those that develop and provide AI-based tools and systems – and do so in the context of the WeVerify project.

Day two, 17 June, focuses on demonstrating the WeVerify tools for countering disinformation. It will begin by introducing the project, followed by demonstrations of three WeVerify tools and the underlying AI technology: Truly Media, the open-source InVID/WeVerify verification plugin,  the WeVerify Database of Known Fakes (DBKF). All registered participants will be provided with a recording of all demonstrations and will be given opportunities to ask questions and register to use the tools post-workshop.

Participation is totally free of charge. All you need to do is register beforehand under this link and be there for a 10 am CEST start on 16 June 2021. Day one will run until 2 pm and include some short breaks in between. So make sure to secure your (virtual) seat now by registering here.

Workshop Programme 

16 June 2021

Host of Day 1: Jochen Spangenberg, WeVerify project / Deutsche Welle

Session 1: Policy implications of technology adoption (10:15 – 11:00)

This session presents how the EU and selected European bodies regard the role of Artificial Intelligence and technology leveraged within civil society to fight disinformation and provides guidance on strengthening the Code of Practice on Disinformation. We will hear from the EU Commission as well as from the European Digital Media Observatory (EDMO), an EU-funded initiative that is considered to  play an important role in countering disinformation and supporting research and related fact-checking activities. 

Discussion Topics and Speakers: 

• Dr. Alberto Rabacchin, Scientific Project Officer, European Commission.  
• Paula Gori, Secretary-General and Project Manager Leader of EDMO

The session will be chaired by WeVerify’s / EU DisinfoLab’s Alexandre Alaphilippe.

Session 2: News from the field: technology and verification in journalism and the human rights sector (11:10 – 12:30)

The session features three distinguished expert practitioners that deal with digital content analysis and its verification on a regular basis. They work in both journalism and the human rights sector. In their talks, they will give insights into their teams’ daily workings and share experiences past and present, also looking ahead. In particular, they will shed light on challenges and constraints in the verification and fact-checking sector and related processes, especially when it comes to using technology and AI-based tools and systems.

We are very happy and honoured to welcome the following speakers to this session:

• Hazel Baker, Global Head of UGC Newsgathering at Reuters / Reuters Fact Check. 
• Sam Dubberley, Head of Evidence Lab at Amnesty International’s Crisis Response Team
• Sophie Nicholson, Deputy Chief Editor, Digital Verification at AFP Agence France Press / AFP FactCheck.

The session will be moderated by WeVerify’s / Deutsche Welle’s Jochen Spangenberg. 

Session 3: WeVerify advances in technology and AI to counter disinformation (12:45 – 14:00)

This session will focus on the new features built by WeVerify partners in the course of the project. Those tools evaluated with fact-checkers in several iterative sessions bring novel enhancements to better use verification tools, to extract meaningful content, and to make use of AI automation.

• Prof. Kalina Bontcheva (University of Sheffield) is the scientific director of WeVerify and has been researching AI methods for disinformation detection and analysis for over 6 years. She will present the WeVerify verification assistant, which seamlessly embeds AI technology for analysing memes, matching claims against an existing database of known fakes, and flagging content from unreliable sources. 
• Dr. Symeon Papadopoulos is head of the Media Verification (MeVer) team and has been investigating AI and data mining methods for multimedia and social media verification tasks during the past years. He will present an overview of the emerging Internet threat of DeepFakes and will describe the effort towards building the WeVerify DeepFake detection service.
• Antoine Grégoire, is OSINT investigator at the EU Disinfolab. He will present  research in analysing Coordinated Inauthentic Behaviour and the elaboration of the CIB Detection Tree. Antoine has studied the criteria used by platforms and non-platforms to define CIBs, and elaborated a hybrid tool to help humans assisted by AI to detect and prove Coordinated Inauthentic Behaviours.

WeVerify’s / AFP’s Denis Teyssou will chair this session.

17 June 2021

Host of Day 2: Prof. Kalina Bontcheva, WeVerify project / University of Sheffield

Session 1: WeVerify Project Introduction (10:00 – 10:20)

Session 2: InVID/WeVerify browser verification plugin (10:20 – 11:00)

A 35 min demonstration of the widely used open-source InVID/WeVerify browser plugin (over 45 thousand users), with emphasis on its new Twitter social network analysis features, the verification assistant, and the meme OCR WeVerify tool. This will be followed by a 15 min Q&A session.

Session 3: Database of Known Fakes (11:15 – 11:45)

A 20 min demonstration of the WeVerify Database of Known Fakes (DBKF) and its search dashboard, followed by a 10 min Q&A session. In order to help professionals quickly identify cases of content appearing in false context and avoid debunking the same stories again and again, WeVerify has created DBKF. Already debunked content is represented using an extended version of the ClaimReview metadata schema and fact checkers can quickly search for similar debunks based on text, image or video. 

Session 4: Truly Media Demonstration (12:00 – 12:30)

We will provide a 20 min demonstration of the Truly Media content verification workbench, which has recently been established as the infrastructure underpinning the European Digital Media Observatory (EDMO). This will be followed by a 10 min Q&A session.

Session 5: Deep fake detection tools (12:30 – 12:45)

This session will demonstrate the WeVerify deep fake detection technology and its integration within the Truly Media platform.

Session 6: Q&A and closing remarks (12:45 – 13:00)

Registration

Participation in the workshop is free of charge and open to all interested parties. Please register online here.