Session
Classroom
Duration (minutes): 75
Format description: A 75-minute open forum in a "classroom" setting is ideal for starting a conversation on an already adopted and available instrument (Council of Europe Guidance Note on countering the spread of online mis- and disinformation through factchecking and platform design solutions in a human rights' complaint manner) because it allows for structured yet open discussion. This format encourages focused interaction, providing ample time to introduce the topic, outline key points, and engage the audience. The classroom setting fosters a collaborative atmosphere, while the raise-of-hands method ensures that all participants have the opportunity to contribute and voice their opinions. This approach helps clarify any doubts, stimulates deeper engagement, and promotes diverse perspectives, enabling a more comprehensive understanding and progression of the conversation. The duration of 75 minutes will be sufficient to analyse topics that need reflection from substantial but also from technical point of view.
This session will address the challenges deriving from the use of Artificial Intelligence tools generating and spreading disinformation and the distinctive threats these pose to democratic dialogue. The quality of public debate is threatened at various level, ranging from false content spreading at large scale, unlikely to be tackled solely by human intervention, to the propagation of false information by individuals who consider it as true and share it in good faith.
Starting with a presentation of the Council of Europe Guidance Note on countering the spread of online mis- and disinformation through factchecking and platform design solutions in a human rights’ compliant manner, participants will discuss practical measures policymakers and stakeholders can take, such as support for fact-checking, platform-design solutions and user empowerment. The session will also examine the role and responsibilities of digital platforms in both the dissemination of false AI generated information and the promotion of quality journalism.
Bringing together media professionals, AI experts, policymakers, and other stakeholder, the conversation will highlight AI’s dual role—both as a potential vehicle for producing and distributing disinformation when misused, and as a tool for enhancing fact-based information and play a positive role in enabling a safe, inclusive and favourable online environment for participation in public debate.
Panellists will share their experiences, challenges, and strategies for combating AI-driven disinformation and the efforts put in place to maintaining trust in news production. The discussion will also address the challenges arising in this context from the rising use of generative Artificial Intelligence systems, including key technologies like deepfakes, highlighting the need for regular updates and careful vigilance in understanding disinformation.
To foster an interactive and inclusive dialogue, audience members — both onsite and online — will be encouraged to ask direct questions and actively engage in the discussion.
The session will conclude with a summary of key takeaways, reinforcing the importance of assessing AI potential benefits and risks in countering the spread of online mis- and disinformation.
Council of Europe
Speaker 1: David Caswell, Product developer, Consultant and Researcher of computational and automated forms of journalism
Speaker 2: Chine LABBÉ, Editor-in-Chief and Vice President of Partnerships at NewsGuard
Speaker 3: Maria NORDSTRÖM, PhD, Head of Section, Digital Government Division, Ministry of Finance, Sweden.
Speaker 4: Olha PETRIV, Artificial intelligence lawyer, Center for Democracy and the Rule of Law (CEDEM), Ukraine
16.10
Targets: Disinformation undermines trust in the media and threatens the reliability of information that feeds public debate. Combating disinformation and ensuring a healthy, plural and reliable digital environment is crucial for upholding democratic values. Addressing the risks posed by AI and expanding the benefits that can derive from its use to enhance transparency, accuracy, and accountability in the information landscape, directly serves the UN goal of ensuring public access to information and protecting fundamental freedoms.
Practical guidance and recommendations to policymakers and stakeholders (including governments, regulators, industry, journalists, civil society, researchers, and users) on countering the dissemination of online disinformation through fact-checking and platform-design solutions, all in a human-rights compliant manner, and with due regard to user empowerment, is an important precondition for addressing the negative impacts of AI on disinformation.
Report
The Open Forum 16 “AI and Disinformation: Countering the Threats to Democratic Dialogue” was organised by the Council of Europe. It enabled participants to explore the pressing issue of countering disinformation in the age of artificial intelligence (AI). It was highlighted that countering disinformation remains a top priority for the Council of Europe, which has recently adopted the Framework Convention on Artificial Intelligence, Human Rights, Democracy and the Rule of Law — the first legally binding international treaty on AI. Additionally, the non-binding Guidance Note on Countering Online Mis- and Disinformation sets out recommendations built around three pillars: fact-checking, platform design rooted in human rights, and user empowerment, providing a solid foundation for future work in this area.
The discussion focused on how AI can both challenge and facilitate the safeguarding of a healthy digital information ecosystem. Speakers reflected on the evolution of disinformation from pre-AI “artisanal” efforts to the current era of large-scale “automated” dissemination. It was noted that AI holds the potential to accelerate the fragmentation of shared narratives across various sectors, including journalism and government intelligence. Key risks include persistent disinformation, at-scale radicalisation, and biases embedded in AI models. The need for systemic, inclusive, and empowering approaches to strengthen the information ecosystem was emphasised.
Speakers also addressed the surge in deepfakes and AI-generated content, highlighting their potentially severe consequences, especially in conflict contexts. Disinformation campaigns were found to be now cheaper and more effective, creating a vicious cycle in which AI-generated falsehoods are further amplified by other AI systems. Reference was made to the phenomenon of “LLM grooming” — a tactic where malicious actors deliberately feed large language models with mis- and disinformation. It was noted that this problem is even more acute in non-English contexts, where chatbots accuracy is lower, and where press diversity is more limited.
The Open Forum provided for the opportunity to also focus on the recently adopted Council of Europe Framework Convention on Artificial Intelligence (adopted on 17 May 2024) and the principles it establishes to safeguard human rights, democracy, and the rule of law, while supporting innovation. Information was provided also on the ongoing work of the Council of Europe Committee on Artificial Intelligence (CAI), which is developing a risk and impact assessment methodology to guide ethical and responsible AI deployment by developers and other stakeholders. Panellists shared relevant national practices to tackle AI-generated disinformation and showcased initiatives designed to enhance AI literacy, empowering citizens to critically assess AI-generated content.
Positive examples included also bottom-up approaches to work with ethical AI, promoting self-regulation and annual transparency reporting. Speakers emphasised the importance of enhancing resilience against AI-generated disinformation starting with education, particularly for children, who are both vulnerable to, and potential unwitting spreaders, of false information. It was noted that early education empowers children to understand algorithms and use AI creatively and critically. Education should be seen as not only a protective measure but as a long-term investment in democratic resilience.
Participants in the Open Forum engaged in the Q&A session, further addressing key issues raised during the discussion, including: the benefits of effective education and strategies to strengthen digital literacy; the role of fact-checkers; the use of AI chatbots to promote media literacy; incentives for providers to prioritise trustworthy information; and co-regulation models that foster truth-seeking as a market incentive.
Overall, the panel underscored the dual nature of AI as both a challenge and a potential ally in the fight against disinformation. Strengthening regulatory frameworks, promoting literacy, and fostering ethical self- and co-regulation were identified as key steps to building a more resilient information environment.
