Deepfakes und manipulierte Realitäten - Technologiefolgenabschätzung und Handlungsempfehlungen für die Schweiz

A deepfake is an audio or (moving) image content synthesised or manipulated with the aid of artificial intelligence technologies, which appears to be authentic and in which a person says or does something that he or she has never said or done. Since 2017, when synthetic and manipulated media content was first
referred to as “deepfake”, the term has become firmly entrenched in political and media debate.

Several years after its introduction, we can now draw some mixed conclusions: some political deepfakes and deepfake-based fraud cases, for example nuisance or scam calls, appear to confirm certain concerns. On the other hand, avariety of useful applications have been created that are based on synthetic and manipulated media content. Neither the widely feared large-scale use of deepfakes in disinformation campaigns nor a major information apocalypse have materialised to date.

The purpose of this study is to assess the opportunities and risks associated with deepfakes for Switzerland.