What happened
Instagram head Adam Mosseri said the platform will not filter out AI-generated content, but will label it and allow users to choose whether to see it. Mosseri acknowledged that detecting AI content is difficult and Instagram may lose the ability to identify AI posts as models improve. He suggested it might be more practical to label non-AI "camera-captured content" instead.
Why it matters
Users frustrated with AI content cannot simply hide it from their feed, though Mosseri believes those who want it should be able to have "a feed that's just AI town." Instagram, like TikTok, YouTube, and Facebook, labels AI-generated content but does not offer filtering options. At the same time, the platform is embracing AI tools—its new image generator, Muse Spark, lets users put other people into AI creations by tagging them, a feature a National Center on Sexual Exploitation executive warned "creates obvious and foreseeable opportunities for exploitation, sexual abuse, harassment, and identity fraud."
What to watch
Mosseri stated Instagram needs to "figure out how to crack down" on spammy AI content, even as the platform continues to expand its AI capabilities and users grapple with distinguishing real from synthetic posts.
Summaries like this, in your inbox every morning.
Instagram's approach to AI content reflects a broader tension between platform openness and user control. Mosseri's position—labeling but not filtering—treats AI as a neutral feature of the platform rather than something users should be able to opt out of entirely. His acknowledgment that Instagram may eventually lose the technical ability to detect AI content suggests the company views labeling as a temporary measure, not a long-term solution. This puts the burden on users to interpret labels and decide what to engage with, rather than empowering them to curate their own feeds.
The rollout of Muse Spark complicates this stance. While Mosseri frames AI tools as value-adds for creators, the image generator's tagging feature has drawn immediate concern from the National Center on Sexual Exploitation, which warned it creates "obvious and foreseeable opportunities for exploitation, sexual abuse, harassment, and identity fraud." This suggests Instagram is moving faster to expand AI capabilities than to develop guardrails around them. Mosseri's statement that the platform needs to "crack down" on spammy AI content acknowledges the problem exists but does not detail how Instagram will do so—or whether it will address harms beyond spam.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
