Skip to content
Submit incident
AII-1002-11569
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.Supporting sources are pending verification.

YouTube Alters Shorts Videos With AI Without Creator Consent

YouTube reportedly ran an experiment using machine learning to automatically enhance select YouTube Shorts videos by unblurring, denoising, and improving clarity without notifying creators. Content creators noticed unnatural visual artifacts including overly smooth skin, sharpened clothing wrinkles, and warped facial features, raising concerns about unauthorized manipulation of their content.

22 August 2025 Β· YouTube Β· Privacy and surveillance
22 August 2025
AII-1002-11569
Quick facts
Category
Privacy and surveillance
Country
United States
Date
22 August 2025
Drafted by
Raidu Incident Detection

What happened

YouTube ran an experiment on select YouTube Shorts that uses machine learning technology to unblur, denoise, and improve clarity in videos during processing without creator knowledge or consent.

Creators including Rick Beato and Rhett Shull discovered unnatural visual artifacts in their uploaded videos, including overly smooth skin textures, sharpened wrinkles in clothing, and warped edges around faces.

Reported impact

Content creators expressed concern that the unnatural appearance could mislead viewers into thinking they used AI in their videos, potentially damaging their reputations.

The lack of transparency and creator control raised concerns about the growing role of AI in mediating information and media consumption in imperceptible ways.

Index analysis

YouTube later acknowledged the experiment and characterized the changes as using "traditional machine learning technology," comparing the process to enhancements made by modern smartphones.

Experts argued there is a significant difference between users controlling AI features on personal devices and a company manipulating content without creator consent, with one researcher dismissing YouTube's distinction between "traditional machine-learning" and generative AI as misdirection.

Governance controls

YouTube stated it would continue to take creator and viewer feedback into consideration as it iterates and improves these features, but initially deployed the experiment without disclosure to creators.

Classification

Category
Privacy and surveillance
Industry
Media, social platforms and marketing
Country
United States
Harm status
Actual harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Reputational
AI systems
YouTube Video Enhancement System

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident