Ana Maria Constantin
Summary: YouTube cracked down on mass-produced AI videos but its fixes are hurting legitimate "faceless" creators who never used AI. The algorithm now favors on-camera humans, causing demonetisation and channel removals based on proxy signals. Creators are scrambling to show faces or change formats while YouTube balances building AI tools and limiting their spread.
Takeaway: Oh, that age old law of unintended consequences… whenever a platform declares it’s banning, or tagging, AI slop, we should always ask how it’s planning on doing that. Those who perhaps were taking the sensible option of preserving their privacy (as I did when I was a old-time mummy blogger) are penalised because of AI content finding them easier to emulate. We really need to think more steps ahead that just one… Crowdsourced ratings, automatic labels and channel-level enforcement risk wrongful demonetisation or bans, while YouTube simultaneously builds AI tools — a conflicted approach that harms privacy-focused, anonymous creators and invites perverse incentives.
YouTube cracked down on mass-produced AI videos but its fixes are hurting legitimate "faceless" creators who never used AI. The algorithm now favors on-camera humans, causing demonetisation and channel removals based on proxy signals. Creators are scrambling to show faces or change formats while YouTube balances building AI tools and limiting their spread.
Highlights
Crowdsourcing AI detection has obvious limitations. Research consistently shows that people are poor at identifying AI-generated content, and their accuracy is declining as the tools improve. There is also no indication of how YouTube will weight the ratings or whether a threshold of negative viewer feedback will trigger demonetisation or suppression.
YouTube is now testing a new approach: a mobile pop-up that asks viewers to rate whether a video feels like AI slop on a five-point scale from “*not at all*” to “*extremely.*” The feature appeared in March 2026 and adds a third layer of detection on top of YouTube’s existing automated and human review systems.
YouTube has a growing AI slop problem, and its efforts to fix it are catching legitimate creators in the crossfire. In January 2026, the platform terminated 16 channels with a combined 35 million subscribers and 4.7 billion lifetime views under its inauthentic content policy, a quiet rename of the old “*repetitious content*” rules. The channels were producing mass-generated, low-effort content at scale, but the algorithm changes that followed are now penalising a much broader group: faceless creators who have never used AI at all.