A million users tap LinkedIn's AI slop button in weeks
Over a million people have tapped LinkedIn's new 'AI slop' flag, weeks after a study found 41% of long posts were machine-written. The signal India's professional feed cannot ignore.
The News
LinkedIn's newest moderation tool is getting a workout. The company says more than one million people have already used its "Seems like AI slop" button, a report option it switched on across the platform on 30 July. The figure was shared this week by chief product officer Hari Srinivasan, who said over a million users had tapped the control in the weeks since launch.
The button sits inside the three-dot menu on any post, giving members a one-tap way to tell LinkedIn that a piece of content reads as low-effort, machine-written filler. It is a crowd-sourced quality signal rather than a takedown request, feeding LinkedIn data on which posts its own users find hollow.
The timing was not accidental. LinkedIn rolled out the flag only weeks after the AI-detection firm Pangram concluded that 41 per cent of the platform's longform posts were flagged as fully machine-generated, a finding first surfaced by the outlet 404 Media. For a network built on professional credibility, that is an uncomfortable number to sit under.
Why It Matters
The million-click milestone is less about the feature and more about what it reveals: users have grown fluent in spotting synthetic writing, and they are irritated enough to report it in bulk. Platforms are now treating AI authenticity the way they once treated spam and misinformation, as a trust problem that can quietly erode engagement if left unchecked.
This echoes the reporting-driven clean-ups of the late 2010s, when Facebook and YouTube leaned heavily on user flags to police low-quality and misleading content at scale. The difference is that the target is no longer obvious spam but polished, plausible prose that a large language model produced in seconds. When two in five long posts trip an AI detector, the risk is a feedback loop where machine-written content trains the next wave of machine-written content, and human readers simply tune out.
For a business-networking product whose entire value rests on signalling genuine expertise, that hollowing-out is an existential threat, not a cosmetic one. The slop button is LinkedIn admitting the problem in public.
Indian Angle
The stakes are sharpest in India, which LinkedIn has repeatedly described as its second-largest market, with a membership base the company has put at more than 130 million. For millions of Indian jobseekers, recruiters and freelancers, the feed is not a novelty but a working tool, and a market where AI-written "thought leadership" has become a booming cottage industry.
Generative writing assistants are cheap and increasingly localised for Indian users, which makes ghostwritten posts easy to mass-produce and hard to police. If recruiters begin distrusting the feed, the people who lose most are the genuine early-career candidates in tier-2 and tier-3 cities who rely on LinkedIn visibility to get noticed without a network. A credibility crisis on the platform hits India's aspirational professional class first.
There is a regulatory shadow, too. MeitY has been tightening expectations around synthetic and machine-made content, and platforms operating at Indian scale will face growing pressure to label or moderate machine-made material. A user-driven slop signal is exactly the kind of lightweight mechanism Indian regulators may come to expect as a baseline.
FAQ
When did LinkedIn launch the button?
The "Seems like AI slop" report option went live on 30 July. Chief product officer Hari Srinivasan disclosed this week that more than one million users had already used it, making it one of the platform's fastest-adopted moderation controls.
Where do I find the flag?
It lives in the three-dot menu on any individual post. Tapping it tells LinkedIn that the content reads as low-effort or machine-generated. It is a quality signal, not an automatic takedown, so flagged posts are not immediately removed.
How much LinkedIn content is AI-written?
The AI-detection firm Pangram found that 41 per cent of LinkedIn's longform posts were flagged as fully machine-generated, a result first reported by 404 Media. LinkedIn has not published its own figure.
Why does this matter for Indian users?
India is LinkedIn's second-largest market, with more than 130 million members by the company's own past disclosures. Trust in the feed directly affects hiring visibility for Indian jobseekers, especially early-career candidates outside the metros who depend on the platform.
This story was reported by The Verge. Read the full original coverage at The Verge.