Instagram's AI Detection Is Failing as Generators Improve, Mosseri Admits

Adam Mosseri, head of Instagram, acknowledged on Lenny Rachitsky's podcast that the platform may "lose the ability" to detect AI-generated posts as generative models advance. He framed this as inherent difficulty rather than a technical oversight. The admission undercuts Instagram's stated strategy of labeling synthetic content instead of filtering it: if detection is already unreliable and worsening, neither users nor the platform can trust the labels to mean what they claim.