Instagram chief says AI posts should be labeled, not banned from feeds
Adam Mosseri said Instagram should account for users’ AI preferences while stopping short of a full filter that blocks AI-generated posts.
By James Whitfield · Staff Writer
3 min read
Instagram head Adam Mosseri said the app should reduce AI-generated material for users who do not want to see it, while rejecting a platform-wide filter that removes AI posts altogether. The comments matter because The Verge reported that Instagram and other major platforms are adding AI labels without offering a direct switch to exclude synthetic posts from feeds.
Mosseri made the remarks during an appearance on Lenny Rachitsky’s podcast. He said he does not think Instagram should remove AI content from the platform, and said the company should instead tell users whether content appears to be made with AI.
Mosseri framed the issue as a matter of feed ranking rather than a ban. According to The Verge, he said people who dislike AI content should not have it in their feeds, while people who enjoy it should be able to see a feed made up of AI posts.
The Verge reported that Instagram, TikTok, YouTube and Facebook already label AI-generated content in various ways. The Verge also reported that those services do not give users an option to filter AI-generated content out of their feeds.
Detection remains a problem
Mosseri said on the podcast that identifying AI-made posts is difficult. He also said Instagram could lose the ability to detect some AI content as generative models improve.
According to Mosseri, Instagram should be able to respond when a user asks whether something is AI-generated. He said the app could answer with degrees of certainty, such as saying it probably is AI, that Instagram is unsure, or that it is definitely not.
Mosseri also suggested that labeling media captured by a camera could become more practical than trying to identify every piece of AI output. The Verge reported that this echoed comments he made in December 2025 about fingerprinting “real media.”
The distinction is significant for Instagram because AI detection can become less reliable as tools improve. Mosseri’s comments point toward a system that gives users more context and uses feed ranking, rather than one that removes an entire category of content from the service.
Instagram keeps adding AI tools
Mosseri said Instagram still needs to address spammy AI content, according to The Verge. At the same time, The Verge reported that the platform continues to add AI features tied to Meta’s generative tools.
One example cited by The Verge is Meta’s AI image generator, Muse Spark. The Verge reported that Instagram users can use the tool to place other users into AI creations by tagging them.
Haley McNamara, executive director of the National Center on Sexual Exploitation, criticized that feature in comments reported by The Verge. McNamara said it creates “obvious and foreseeable opportunities for exploitation, sexual abuse, harassment, and identity fraud.”
Mosseri’s position leaves Instagram in a middle ground: AI posts can remain on the platform, labels can help users understand what they are seeing, and feed systems can respond to individual preference. According to The Verge, Instagram has not offered a full AI-content filter despite growing calls from users who want one.
This story draws on original reporting from The Verge.