YouTube is Using AI Edits on Videos Without Telling Users

YouTube has quietly begun enhancing creator videos by applying artificial intelligence to Shorts without informing or obtaining consent from content creators, according to a BBC News report.

Several creators and observers have noted subtle yet unexpected changes, prompting an outcry over transparency and content authenticity. Some have found that their videos appeared smoother, sharpened or unexpectedly detailed when uploaded as Shorts.

For example, creator Rhett Shull compared his video side by side with the version on Instagram and found the Shorts version had an unexpected oil painting effect over his face. He stated that he had not agreed to any enhancement and said that such unpermitted changes undermine audience trust in his work and trust in YouTube itself.

Other creators have voiced similar concerns. Music educator Rick Beato, with millions of subscribers, observed that his appearance in a Short looked subtly edited. He described unnatural facial smoothness and slight distortions that he did not initiate.

Online discussions going back to June, including a Reddit thread, show that many noticed wrinkles vanishing, fabrics looking too sharp, or small details like ears appearing warped. These differences are often so subtle they require direct comparisons to detect but are enough to spark unease.

YouTube responded to the concerns by confirming that it is conducting an experiment applying traditional machine learning, not generative AI, to selected Shorts. These algorithms aim to unblur, reduce noise, and improve clarity much like enhancements in modern smartphone cameras. The company emphasized that this is part of an ongoing effort to improve viewer experience.

(Image by Serenity Strull/ Getty Images)

Despite this explanation creators remain uneasy, arguing that even minor automated edits can erode the authenticity of their content. “If I wanted this over-sharpening I would have done it myself,” Shull declared.

Critics say the lack of notification sets a worrying precedent. Altering user uploads behind the scenes could strain creator platforms and trust.