What Is a Deepfake? A Plain-English Guide
Deepfakes explained simply: how they're made, why they've gotten so convincing, and the practical ways to tell a synthetic face or voice from the real thing.
A deepfake is media — usually a video, image or voice clip — where a machine-learning model has swapped, generated or altered a person so it looks or sounds like they did something they never did. The name is a blend of deep learning and fake.
A few years ago deepfakes were a novelty. Today they’re a few taps away, and the best ones are genuinely hard to catch by eye. Here’s what’s actually going on.
How deepfakes are made
Most fall into three buckets:
- Face swaps — one person’s face mapped onto another’s body in a video.
- Full generation — a person (or a whole scene) created from scratch by a diffusion model. Nobody filmed anything.
- Voice cloning — a few seconds of audio is enough to synthesise someone saying anything.
The models learn from large datasets of real faces and voices, then generate new frames that are statistically plausible — which is exactly why they fool us.
Why they’ve gotten so good
Two things changed: the models got dramatically better at fine detail, and the tools got easy. What used to need a GPU and a weekend now runs in an app. The classic giveaways — flickering edges, dead eyes, garbled hands — are mostly gone in current-generation output.
How to protect yourself
You can’t reliably eyeball modern deepfakes, so lean on process instead:
- Consider the source. Where did this actually come from?
- Look for provenance. Content Credentials (C2PA) are a signed receipt of origin — their presence is a strong signal for authenticity.
- Run a forensic check. Tools that fuse multiple signals — metadata, face analysis, frequency patterns — catch what your eyes can’t.
That last point is the whole idea behind Verifyco: a fast, on-device second opinion before you trust — or share — a piece of media.
