Instagram won't filter AI content, but lets users opt out of feeds

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.