27/08/2026
When does AI-generated or AI-manipulated content become a deepfake?
In the second instalment of the FBE’s “Deep Fakes in the Courtroom” series, we move from definition to practical examples. A manipulated photograph, an altered video, synthetic voice cloning, an interactive avatar or a digitally recreated context may all qualify as deepfakes where the content is constructed or presented in a way that makes it appear authentic or truthful.
The crucial distinction is not merely technical. Not every piece of AI-generated content is a deepfake. Fantasy, satire, fiction and artistic creations generally remain outside this category when they are recognisable as such and do not purport to represent real facts, persons, statements or situations.
The key question is therefore whether the content may lead its recipient to believe that it represents something real. This distinction is particularly important for lawyers who may encounter synthetic or manipulated material as evidence, in client communications or in court proceedings.
Discover the second one-pager prepared by the FBE New Technologies Commission and follow the FBE’s social media channels for the next instalments of the series.
DigitalEvidence ArtificialIntelligence AccessToJustice
