Safety
Deepfake
Synthetic media — fake video, audio, or images of a person generated with deep learning; a key misinformation, fraud, and safety concern in 2025-2026.
Deepfake is synthetic media — a video, audio clip, or image in which a person’s likeness or voice is falsely generated or altered using artificial intelligence, with a realism that makes it hard to distinguish from authentic content. The name blends “deep learning” and “fake,” and the term emerged in late 2017.
Technically, deepfakes rely on deep learning: early versions used autoencoders and generative adversarial networks (GANs), while modern systems increasingly use diffusion models for face synthesis, face swapping, and voice cloning. A generator produces the fake content, and quality keeps rising as synthesis techniques outpace detectors.
Through 2025-2026, deepfakes rank among the top AI safety concerns — political disinformation, financial fraud, fake celebrity endorsements, and non-consensual content. Regulators respond with labeling requirements: Article 50 of the EU AI Act mandates that artificially generated or manipulated content be clearly disclosed.