X is now automatically labeling photos that are fully or partially generated by AI - meaning images you post that were created with tools like DALL-E 3 or edited with AI features in apps like Adobe Firefly will carry a visible marker visible to anyone who sees the post.
The rollout covers both fully AI-generated images and photos where AI was used to alter part of the image - think AI background removal, object insertion, or generative fill edits. The threshold between "AI-assisted" and "AI-edited" matters here, and X hasn't published a detailed breakdown of exactly which editing operations trigger the label.
For most content creators and marketers using AI visuals on X, the practical effect is a visible disclosure they didn't opt into. That's a meaningful shift. Until now, posting AI-generated images on X was identical to posting a photograph - no flag, no label, no indication to the audience. That era is over.
Whether this hurts engagement is the open question. Audiences on X have very different reactions depending on context. A brand posting AI product shots may see backlash; a digital artist posting clearly stylized AI art may see no change at all. The label reads as a transparency signal in one context and a credibility hit in another.
Meta has had similar AI labeling policies in place on Instagram and Facebook since mid-2024, so X is catching up to an industry norm rather than setting one. The broader direction is clear: platforms are moving toward mandatory disclosure for synthetic media, and the expectation that you can quietly post AI content without any marker is increasingly a thing of the past.