Deep fake (machine learning)

A deep fake is any audio, visual, or audio-visual presentation that is generated using machine learning algorithms (artificial intelligence), utilising a facsimile model of a real person or setting to imitate said real person or setting, without the direct involvement of that person. The rapid and public development of this technology has enabled more and more convincing "fake" audio and video to be produced by a growing number of individuals, requiring only access to now widely available computer hardware and free, open-source software, to the point that it may now be said to undermine the credibility of genuine audio and video, radically reducing the trustworthiness of recorded data in public life.