πŸ“– what

What is a deepfake?

TechnologyArtificial Intelligence

Quick Answer

A deepfake is a video, image or audio clip created or altered with artificial intelligence to make someone appear to say or do something they never did.

The Full Story

Deepfakes use deep learning, a type of AI trained on large amounts of photos, video or voice recordings of a person, to generate realistic fakes. The term came from a Reddit user called deepfakes in 2017, building on generative adversarial networks, a technique introduced by Ian Goodfellow in 2014. Early viral examples included a 2018 public service video in which Jordan Peele voiced a fake Barack Obama. The risks are serious: in 2024 an employee at engineering firm Arup in Hong Kong was tricked into transferring about 25 million US dollars after a video call with deepfaked colleagues. Laws are catching up, including the US TAKE IT DOWN Act of 2025 targeting non-consensual intimate deepfakes, and the EU AI Act requires AI-generated content to be labelled.

Key Facts

1.The word deepfake appeared on Reddit in 2017.
2.A deepfake video call scam cost engineering firm Arup about 25 million dollars in 2024.
3.The US TAKE IT DOWN Act was signed into law in 2025.

YouTube Angle

Suggested video title for this topic:

"Can You Spot the Fake? How Deepfakes Are Made and Detected"