There is a joke buried in this week’s news, and every outlet that covered it reached for the same punchline: the first major social platform to ship a dedicated “seems like AI slop” report button is LinkedIn — the site whose feed practically invented the genre. It’s easy to laugh. It’s more interesting to ask why this platform, and why now.
Judging the vibe, not the origin
The telling detail is the label. LinkedIn didn’t build a “this post is AI-generated” button; it built a “seems like AI slop” one. That phrasing is a quiet admission that reliable AI detection remains a fantasy — you cannot prove a post was machine-written, so LinkedIn is instead crowdsourcing a judgment about quality. Slop isn’t defined by its origin but by its texture: the em-dash cadence, the empty inspirational arc, the confident nothingness. By outsourcing that call to readers, LinkedIn sidesteps the detection problem entirely and turns “slop” into a community-defined vibe rather than a forensic fact.
That’s clever, but it’s also a liability. A subjective flag is a subjective weapon. Human-written posts that merely read like AI — the earnest, the formulaic, the non-native English speaker — will get swept up. The button measures perceived polish, and polish is not the same as authorship.
The part that actually matters
The headline everyone ran with is the button. The more revealing item is buried lower: LinkedIn is retiring its own AI writing feature and replacing it with a proofreading tool. This is the story. A Microsoft-owned platform, sitting inside the company most aggressively shoving generative AI into every text field it owns, has looked at its feed and concluded that helping users generate posts made the product worse. Downgrading from “write this for me” to “fix my typos” is a rare corporate concession that more AI was the problem, not the solution.
Read together, the two moves form a coherent thesis: the value of a professional network is signal, and generative tools flooded it with noise faster than engagement metrics could hide the damage. LinkedIn’s currency is the belief that a real person with a real reputation is on the other end. Slop is corrosive there in a way it isn’t on an entertainment feed, because it poisons the one thing the platform sells.
Whose job is moderation now?
What LinkedIn is really doing is drafting its users as unpaid moderators of an AI mess that platforms — LinkedIn very much included — created and profited from. The button is a pressure valve that converts frustration into free labor and reframes a platform failure as a community feature. It may even work at the margins.
But a report button treats the symptom. As long as posting slop is frictionless and rewarded by reach, no amount of flagging closes the gap. LinkedIn deserves credit for being first to admit the problem out loud. The test is whether it’s willing to touch the incentives that manufacture it.