X may start nudging users to label posts it thinks were made with AI

It seems like X is developing some kind of automatic detection for AI-generated content, and it could show a prompt right when you try to post something it flags. A leaker found strings in the app that suggest a notification on the compose screen saying “AI generated content detected,” popping up before the post actually goes out.

This comes from app researcher Aaron, who goes by @aaronp613. He shared a screenshot of the string along with a key called COMPOSE_VIEW_C2PA_AI_DETECTED_TOAST. The “TOAST” part in that key stands out. In mobile app terms, a toast is just a light, temporary notice that doesn’t block you. So it feels more like a gentle reminder than a full barrier, maybe encouraging folks to use X’s “Made with AI” label instead of stopping the post cold.

X began introducing that label last month, letting users mark their posts as AI-generated or edited if they want. If this new detection rolls out, it would build on that by spotting AI stuff on its own, rather than waiting for people to flag it themselves.

The system looks to rely on C2PA, which is a standard supported by a group of companies like Adobe, Microsoft, and Google. It’s meant to trace where content comes from and highlight anything made or tweaked by AI. But how well X’s version works on a big scale is up in the air. AI detectors are known for messing up with false positives, and that could hurt both users and X’s reputation.

Someone in the comments even pointed out how words from the Bible were also apparently detected as AI-generated text.

That said, Aaron’s choice of words — “show a warning” — has left many users puzzled, considering how hard xAI has been prompting Grok lately. Commenters say it’s weird for a platform pushing AI creation to then turn and flag that same content before posting.

This is why I believe that the “TOAST” won’t be a “warning” as such, but just a reminder for users to use the “Made with AI” label, as noted above. Of course, take all this with a grain of salt since we’ll only know how it will actually function when the feature goes live or gets leaked in action.

Interestingly, when the same post was reshared by @xDaily, comments under that post were somewhat positive, noting it’s a steap in the right directly.

Meanwhile, X has been cracking down harder on hidden AI content. Back earlier this month, they announced they’d cut off revenue sharing for creators posting unlabeled AI-generated war footage. The company’s product head, Nikita Bier, also highlighted an instance where the platform found one person using multiple accounts to post fake AI videos, hoping to fool the masses.

None of this is official or active right now. But given how quickly X has been shipping new features, I wouldn’t be surprised if we see the feature in action within the next few weeks, if not earlier.

Featured image generated with AI

We stand out from the tech-media crowd because we break news stories; we mainly bring you stuff that you won’t find anywhere in the mainstream tech media. Our stories have been picked up by some of the world’s most popular websites and media outlets—more info is available here.