The project defalsi-AI aims to identify false information with the help of artificial intelligence. The focus is on the development of media-forensic tools, which are due to be utilised by a wide user base. The project especially addresses politically-motivated false information, which can and often does undermine trust in democracy.

The main research focus of defalsif-AI (detection of false information via artificial intelligence) is on audio-visual media forensics, text analysis, and the fusion of these technologies with the help of AI. A lot of attention is paid to the transparent and interpretable presentation of results – the final decision if contents are “wrong” or not still lies with the users.

The goal is to demonstrate a valid tool for the analysis of digital contents online which allows a first evaluation of texts, images, videos or audio clips for validity and authenticity. The creates the groundwork for further action.

The project also includes an analysis and assessment of the of media-forensic tools from the legal and sociological perspective. Defalsif-Ai is rounded off with the derivation of application oriented, technical and organisational measures as well as a plan of usage for the future use of legally justified disinformation analysis platforms.

The project is funded by the ministry for Finance (BMF) within the KIRAS Security Research programme, and carried out by the Austrian Research Promotion Agency (FFG).

The project is funded by the ministry for Finance (BMF) within the KIRAS Security Research programme, and carried out by the Austrian Research Promotion Agency (FFG).