Instagram’s initiative to display visible AI labels was designed to allow users to instantly identify synthetically generated media. However, recent feedback indicates the system is malfunctioning significantly. Users have noted that Meta is automatically tagging ordinary images with an “AI Content” label despite them not being created or modified via generative artificial intelligence tools.
Conversely, genuine AI-generated imagery is frequently escaping detection. This dual failure has led to a broader sense of distrust regarding the authenticity of visual content on the platform.
Reports suggest the root causes differ across cases. A common complaint involves photos edited using standard design utilities, such as Canva’s background removal feature, which are being incorrectly flagged as AI-generated. The inconsistencies in the current algorithm leave many questioning the reliability of the labeling system.
My travel photos got labeled AI despite being taken with my phone. The system is clearly broken.
Wait, so now I can’t trust the labels? If real photos are flagged and AI slides through, what’s the point?
Canva users have been complaining about this for months. Background removal should not equal AI generation.
This feels like a rushed rollout. Meta needs to fix the accuracy before expanding to other platforms.
I work in digital forensics and this inconsistency is a nightmare for verification workflows.