Instagram Head Adam Mosseri Expects Worse AI Backlash, Says AI Posts Probably Rank Lower

Adam Mosseri‘s forecast that anger at artificial intelligence will keep growing made the headlines. The more useful sentence from his interview was a question he answered himself. The head of Instagram said the app does not rank AI creators differently, then asked: “Do we rank AI content as well as we rank non-AI content? Probably not.”

Those two statements only fit together one way. Instagram does not punish the label on an account; it lets the ranking system judge the post, and by Mosseri’s own estimate AI-made posts tend to lose that judgment. For anyone publishing on the app the difference is mostly semantic. A post that ranks lower reaches fewer people, whatever the policy page calls it.

What Mosseri said about the AI backlash

Mosseri was on the Colin and Samir podcast to talk about Instagram’s move into widescreen, long-form video when the conversation turned to AI and to college students booing tech CEOs at graduation ceremonies, as The Verge reported. He acknowledged that plenty of people dislike AI, not only the young or creatives, and described “more anxiety and more frustration and more anger about AI,” particularly in creative industries.

“It’s not only getting worse, it’s going to continue to get worse,” he said. “I think we’re going to see more frustration and anger, some of it founded, some of it unfounded, over the next couple of years, and that’s going to make it hard for a lot of people, not just in the industry.”

On the ranking question he drew the distinction himself: “There’s a lot of great AI content out there, there’s also a lot of AI slop.” He also said the word “slop” gets used “too liberally”, a term he had already raised in a New Year’s slideshow post. And he said Instagram is “now starting to require AI accounts to label themselves, because we think it’s important that people know if what they’re looking at is real or not real.”

The one AI penalty Instagram has written down

The only explicit AI penalty on Instagram is about disclosure, not quality. Accounts built around an AI-generated person now have to carry an “AI-generated profile” label, which replaced the old optional “AI creator” tag. Accounts that skip it “may soon see limits to their reach,” an Instagram spokesperson told Engadget, with appeals handled through the Account Status dashboard. Engadget called it “less of a crackdown on all AI slop” than a rule for one especially misleading kind of content, and Business Today reached the same conclusion.

That leaves Instagram with two levers. The written one says disclose or lose reach, and it applies to a narrow class of accounts. The unwritten one is the ranking system’s view of quality, and it covers every post on the app. Mosseri’s “probably” gives that second lever a direction for AI content, but not a size.

From grievances to a feeling

Mosseri also listed what worries people. They are concerned, he said, “about owning their content and models getting trained on their content,” about safety, and about job displacement, “if you’re a content creator or an engineer or a mathematician.” Then he folded the list into one diagnosis: “fundamentally what I think it comes down to when you really have, like, a heart-to-heart with someone is a fear of being left behind in some way.”

The items on his list and his conclusion are different kinds of claim. Training on someone’s work and losing income are questions of ownership and money, while “a fear of being left behind” is a mood. A platform can answer a mood with new tools and reassurance. Ownership questions need consent and payment. Mosseri did not say which complaints he files under “founded” and which under “unfounded.”

Why the setting matters

The interview was part of the rollout of Features, Instagram’s widescreen long-form format, which Mosseri presented in a video with Tessa Lyons, Instagram’s VP of product. “For a long time, I have been against doing long-form video on Instagram,” he said there, “but we are actually going to lean in for a bunch of different reasons.” According to PetaPixel, Features supports videos of up to 90 minutes, is a limited test for some creators, and has no long-term monetization scheme yet. Android Central reports that more pieces will arrive in testing over the next month, while Reels stay “at the core” of the app.

So Mosseri was courting long-form creators, a group his own list names among the worried, while saying their AI-made competition probably ranks below them. Engadget’s description of the parent company sits awkwardly next to that pitch: Meta, it wrote, “often encourages users to make their own slop” and “has done very little to discourage AI content.”

The Verge published the remarks on October 28. Instagram’s last long-form attempt, IGTV, launched in 2018 and shut down in March 2022.

For a creator weighing whether to post AI-made work on Instagram, the forecast about rising anger changes little. The word “probably” matters more. It came from the head of the app rather than from a policy page, and it tells creators which way the feed leans on AI work without telling them how far.

Tags: artificial intelligence, meta, instagram, Adam Mosseri, social networks