When you flag a LinkedIn post as 'AI slop', does your feed really get cleaner, or has nobody measured it?

This asks whether flagging AI posts on a platform like LinkedIn actually leads to fewer of them showing up in your feed. The short answer is that nobody has measured it yet, and the corpus gives reasons to think it may work less well than it sounds. LinkedIn has added a 'Seems like AI slop' option to its post menu and says flagged content will get less distribution Will LinkedIn's AI slop flag reduce AI-generated posts?. It also says the reports will train classifiers that reduce slop across the platform Does LinkedIn's AI slop button actually reduce low-quality content?. Neither announcement gives accuracy figures, data on how often the button gets used, or any before-and-after look at what members see. LinkedIn's separate private flags for writers who lean heavily on AI also come with no evidence that those writers change anything Do private AI flags actually change how people write?.

The bigger problem is what those flags would teach a classifier. A study of 25 million Hacker News and Reddit comments found that the writing features that actually separate AI text from human text don't predict which comments people accuse of being slop Do AI slop accusations actually detect AI text?. The 'slop' label works more like social policing (marking what a community finds off-key) than like detection. A classifier trained on member flags may learn what annoys people rather than what a machine wrote. Automated detectors have a related blind spot. Fake news detectors mistake the way LLMs write for signs of deception, so they flag truthful AI text and miss human-written disinformation Why do fake news detectors flag AI-generated truthful content?. Learning a surface style is not the same as identifying where a text came from.

The flag also works against the feed's own incentives. On Reddit, machine-generated comments get about as much engagement as human ones, sometimes more Does machine-generated text get penalized in online engagement?. On Medium, AI-labeled posts got roughly half the likes of human posts, and even that study called the gap small Do readers engage less with AI-generated social media posts?. Polished, thorough AI posts collect likes without starting conversations Why do AI posts get likes without inviting conversation?. So a ranking system built around engagement can keep pushing them up even while a minority of members flag them down. And the flag is pushing against a rising tide: on Medium and Quora, the share of posts a detector attributes to AI went from about 2% to about 38% in under three years Is AI-generated content rising faster on some platforms?, and roughly a third of new websites are now AI-generated or AI-assisted How much of the internet is AI-generated now?. Even if your flags tune your own feed, the pool they filter from keeps getting bigger.

The less obvious point is that 'seeing fewer AI posts' may not be what readers care about. One line of argument holds that the real harm of AI content is the loss of conversation itself: posts that address no one and invite no reply Does AI threaten social media's conversational function?. In this view, the damage happens below the level that moderation or ranking tweaks can reach. AI content also takes attention away from human voices without building any speaker's lasting reputation Does AI content displace human influencers on social media?. A flag that suppresses some obviously generic posts could leave that underlying shift in place. The evidence that would settle the original question, measured exposure for flaggers before and after they flag, isn't in the corpus.