The media sector is exposed to and undergoing continuous innovations that occur at a pace never seen before and have a non-negligible impact on citizens, democracy and a society as whole. A significant booster becomes a generative Artificial Intelligence, which already plays and will continue to play a critical role (in a positive as well as negative meaning) in creating and spreading information. Especially in next-generation social media, which refer to the anticipated evolution towards more AI-based decentralised and immersive virtual environments (like fediverses and metaverses), generative AI can become the most prominent enabler of disinformation growth accompanied by a lack of trusted information. Media professionals are not, however, currently well-equipped with supporting tools nor knowledge to operate in such already emerging environments. As a result, there is a tremendous need for innovative (AI-based) solutions ensuring media freedom and pluralism, delivering credible and truthful information as well as combating highly disinformative content. The main goal of the AI-CODE project is to evolve state-of-the-art research results (tools, technologies, and know-how) from the past and ongoing EU-funded research projects focused on disinformation to a novel ecosystem of services that will proactively support media professionals in trusted information production through AI. First, the project aims to identify, analyse, and understand future developments of next-generation social media in the context of rapid development of generative Artificial Intelligence and how such a combination can impact the (dis)information space. Second, the project aims to provide media professionals with novel AI-based services to coach them how to work in emerging digital environments and how to utilise generative AI effectively and credibly, to detect new forms of content manipulation, as well as to assess the reputation and credibility of sources and their content.

The purpose of this project is to protect democracy in the EU by strengthening the ability to deny the intended effects of FIMI on society. This will be achieved with focused research that brings together some of the principal actors behind the intellectual and technical components of FIMI as it has been developed by the EEAS and other EU Institutions. We will work together to significantly develop upon current knowledge of how FIMI can be detected, categorised, analysed, shared, and countered. We will achieve this through a series of coordinated contributions to the DISARM Framework, the NATO-Hybrid COE Attribution Framework, STIX 2.1, OpenCTI, ABCDE, and the FIMI countermeasures toolbox. This approach acknowledges the importance of TTPs and common data handling standards to the ability to attribute FIMI actors, and further positions TTPs within the broader analytical processes that are necessary to developing countermeasures. In addition to establishing improved technical standards and operating procedures, we will generate research knowledge that can support better decision-making about FIMI countermeasures. For example, we will conduct research into the public impact of attribution, research methods for linguistic and visual analysis, develop the understanding of how cross-platform manipulation evades traditional analysis methods, as well as establishing a dataset of previous FIMI interventions. We include a specific component on gendered disinformation designed to better integrate gender into the technical formats. Finally, we will work closely with a community of practice that includes the EEAS, representatives of member states, civil society, and journalists/EDMO.

Although the European Union's disinformation policy has significantly developed since 2015, Russia’s war in Ukraine has reminded us all of the urgent need to stay ahead of the curve in order to fortify our resilience against the ever-evolving threat of Foreign Information Manipulation and Interference (FIMI). In the last year, Russia’s strategic use of FIMI in its war of aggression against Ukraine has brought an unprecedented response by those defenders who seek to safeguard universal values, democracy, and freedoms by exposing Russia’s efforts to manipulate global discourses. While governance, law, protocols, data exchange, and forensic practices should converge to address FIMI, they currently remain disparate and unaligned. This fragmented approach creates a significant hurdle in the fight against FIMI, hindering the ability of member states to prosecute, ban, attribute, or sanction FIMI actors or campaigns originating from state and non-state actors. To overcome this hurdle there is a need to better synchronise the interventions and harmonise our approach towards FIMI, as proven in the case of cyberattacks. Our consortium, SAUFEX (Secure Automated Unified Framework for Exchange), comprises of concerned stakeholders who possess significant expertise and hands-on experience with the current systems and structures concerning FIMI and are situated in the EU's Eastern regions, where FIMI incidents are frequently observed. We strongly understand the importance of consolidating and utilising available information to establish a collective comprehension of FIMI. To achieve this, our project proposes the creation of a FIMI data-space, which we will herby refer to as a FIMI knowledge database, as well as the establishment of a Resilience Council to act as an intermediary between state and non-state actors, specifically regarding interventions and FIMI attribution.