DisAI-AMPLIFIED: Progress in Multimodal Prediction and the Use of Interactive Fact-Checking to Support Disinformation Combatting

We are thrilled to announce that KInIT has received support for DisAI-AMPLIFIED, a significant extension of our ongoing Horizon Europe DisAI project. With DisAI-AMPLIFIED, we are excited to push the boundaries of AI in disinformation combatting by leveraging the most recent advancements in natural language technologies, helping to safeguard the digital space.

The DisAI-AMPLIFIED project builds on the research component of the ongoing Horizon Europe DisAI project, which aims to enhance the scientific excellence of the Kempelen Institute of Intelligent Technologies (KInIT) and its partners in the field of trustworthy artificial intelligence for combating disinformation. Since the inception of DisAI, natural language processing (NLP) has advanced significantly, particularly in its shift towards the use of large language models with conversational capabilities and support for the vision modality.

DisAI-AMPLIFIED extends the original project in line with these recent advancements. Specifically, DisAI-AMPLIFIED will explore the integration of conversational AI to enhance fact-checking by enabling a deeper, contextual understanding of disinformation. The project will also investigate the use of multimodal models to incorporate visual content, which is essential for effectively addressing misinformation in contexts where text alone is not sufficient. By extending the original DisAI project’s scope in these directions, DisAI-AMPLIFIED aims to create a more comprehensive approach to combating disinformation, utilizing the full potential of advanced NLP and multimodal AI.