Skip to content

AI

LinkedIn Adds a "Seems Like AI Slop" Button. Humans Finally Get a Vote

LinkedIn now lets users flag posts that look machine-written. Substack and Pangram are working the same problem, and Pangram just banked $9 million to do it.

TL;DR

  1. LinkedIn adds a 'seems like AI slop' button so users can flag suspect posts.
  2. Substack added an AI-detection tool via Pangram; Pangram raised $9 million.
  3. Cloudflare says bots now outnumber humans in web requests. Sapiens, we are the minority.

What happened

On Thursday, LinkedIn announced a new feature that lets users click a "seems like AI slop" button when a post looks like it was written with AI. The target is low-quality, artificially generated content clogging the feed.

It is a flag from readers, not a lie detector. The button turns every user into a tiny, unpaid content inspector, which is a very LinkedIn way to solve a LinkedIn problem.

LinkedIn is not alone. Last week, newsletter platform Substack added a tool that helps readers spot when what they are reading was written by AI. It works through a partnership with Pangram.

Pangram, for its part, announced $9 million in new funding this week to tackle AI content flooding the internet.

The pain is not limited to the big names. Per TechCrunch, Digg shut down its attempt at a Reddit competitor in March, saying it could not get a handle on the number of bots flooding the site.

And Cloudflare, an internet infrastructure firm, says the problem keeps getting worse. By its count, bots now send more web traffic than people do, and that moment came sooner than it expected.

Why it matters

Platforms spent years begging people to post more. Now they are building tools to catch posts nobody really wrote. The incentives flipped, and the feed is the evidence.

The LinkedIn button is incremental, not a revolution. It relies on human judgment, and human judgment about what "sounds like AI" is famously shaky. A tidy, well-structured post can be flagged just for being tidy.

Still, the direction is clear. Three separate moves in about a week, from LinkedIn, Substack and a freshly funded Pangram, show that authenticity is becoming a product feature rather than an assumption.

Platforms now sell proof that a human was involved.

For creators, that changes the math. If readers can flag you, copy-pasting a chatbot draft with a motivational opener carries a reputational cost it did not carry before.

What to do with it

  • Edit like a human.If AI drafts your posts, rewrite them with your own details, numbers and opinions before publishing.
  • Skip the template voice.The fastest way to look like slop is to sound like every other post in the feed.
  • Use the button sparingly.Flag the obvious copy-paste spam, not the colleague who just writes neatly.
  • Watch the detectors.Try Substack's Pangram-powered tool if you publish there, and treat any detection score as a hint, not a verdict.

Ciao, sapiens. Write something only you could have written; the bots are busy outnumbering us.