Instagram won't filter AI content, but Mosseri says users should control their feeds

Instagram head Adam Mosseri says users who don't want to see AI-generated content shouldn't have it in their feeds — but the platform won't give them a way to filter it out completely.

In an interview on Lenny Rachitsky's podcast, Mosseri outlined Instagram's approach to the growing volume of AI content on the platform. "I don't think we should filter out AI content," he said. "I think we should let you know if content is AI content or not."

The distinction matters. Mosseri envisions a system where users can sort their feeds based on content type rather than blocking AI posts altogether. He suggested that people who enjoy AI-generated material "should be able to have a feed that's just AI town," while those who prefer human-created content could theoretically curate accordingly.

The detection challenge

Mosseri acknowledged that identifying AI content is becoming increasingly difficult as generative models improve. Instagram may "lose the ability" to reliably detect AI posts, he said, as the technology advances.

His proposed solution involves a more interactive approach: users should be able to ask "Is this AI?" and receive an answer indicating whether Instagram thinks content is probably AI-generated, uncertain, or definitely human-made.

Interestingly, Mosseri suggested it might be "more practical" to label non-AI content instead — specifically marking "camera-captured content" as authentic. This echoes his December 2025 comments about fingerprinting "real media" rather than trying to catch every AI-generated post.

Policy versus practice

Instagram currently labels AI-generated content, following the lead of other major platforms including TikTok, YouTube, and Facebook. But none of these services offer users the option to filter AI posts out of their feeds entirely.

Meanwhile, Instagram continues expanding its AI features. Meta's AI image generator, Muse Spark, now allows users to create AI images featuring other Instagram users simply by tagging them. Haley McNamara, executive director of the National Center on Sexual Exploitation, warned this feature "creates obvious and foreseeable opportunities for exploitation, sexual abuse, harassment, and identity fraud."

Mosseri said Instagram needs to "figure out how to crack down" on spammy AI content, though he provided no specifics on enforcement mechanisms or timelines.

Why it matters

The tension between labeling and filtering reflects a broader platform dilemma: how to manage AI content without either alienating users who create it or overwhelming those who want to avoid it. As AI generation becomes harder to detect and easier to produce, platforms that rely on algorithmic feeds face mounting pressure to give users more granular control. Instagram's current stance — supporting labels but not filters — may prove untenable if users can't effectively curate their experience or if safety issues around AI-generated content escalate.

These details were first reported by The Verge.