"Over the course of 2025, deepfakes improved dramatically. AI-generated faces, voices and full-body performances that mimic real people increased in quality far beyond what even many experts expected would be the case just a few years ago. They were also increasingly used to deceive people. For many everyday scenarios - especially low-resolution video calls and media shared on social media platforms - their realism is now high enough to reliably fool nonexpert viewers."
"Several technical shifts underlie this dramatic escalation. First, video realism made a significant leap thanks to video generation models designed specifically to maintain temporal consistency. These models produce videos that have coherent motion, consistent identities of the people portrayed, and content that makes sense from one frame to the next. The models disentangle the information related to representing a person's identity from the information about motion so that the same motion can be mapped to different identities,"
Deepfakes improved substantially in 2025, achieving realism that often fools ordinary viewers in low-resolution video calls and social media. The total number of online deepfakes rose explosively, with an estimate increasing from roughly 500,000 in 2023 to about 8 million in 2025, near 900% annual growth. Technical advances centered on video-generation models that maintain temporal consistency and separate identity representation from motion, producing stable, coherent faces without prior artifacts. Synthetic media are now broadly accessible, enabling almost anyone to create deepfakes, and the trend points toward more interactive, real-time synthetic performers in 2026.
Read at Fortune
Unable to calculate read time
Collection
[
|
...
]