“Seems like AI slop” LinkedIn’s new feature and what it means for you

Last week LinkedIn shipped a pretty gutsy feature. You can now report posts for sounding like AI. Click the three dots on any post and you now have the option to report it as “Seems like AI slop”

We’ve seen stuff like this before, YouTube’s dislike button, Spotify’s “Don’t include this in my taste profile” feature. What made this unique is the lack of preamble. One day we were all lamenting the seemingly endless sea of AI generated posts and the next, here is a new bit of functionality specifically designed to tackle the issue. 

What LinkedIn said

During his announcement video, LinkedIn’s CEO Daniel Shapero specifically said that he does not think that AI-generated “AI slop” has a place in the feed. They are doubling down on the idea that the feed should be driven by insight and personality. Shapero was very clear, AI generated content is going to be penalised. I suspect this reporting tool is the tip of the iceberg and that LinkedIn will continue to outwardly, and algorithmically, clamp down on AI generated content. 

“I know that starting from a blank page and generating insights is hard, but there is just no shortcut to figuring out something unique to say that’s going to be valuable on the platform.” Daniel Shapero, CEO LinkedIn. 

Creators will be given private feedback if their posts are flagged as AI slop, it will show in their analytics dashboard. So, thankfully no public shaming. It’s unclear though whether those posts will be suppressed or penalised in the feed in any way. I suspect they will. 

What does it mean for you?

Well, it depends. LinkedIn has been clear that they define AI and AI slop differently. Dan Shapero encouraged members to use AI for proofreading and “tightening” posts but not for out and out creation. The trouble is LinkedIn and its members define slop differently. LinkedIn knows that many of their members use AI to refine their thoughts, so they have framed the button to invoke an emotional gut feeling “Does this feel inauthentic?” rather than “Was AI directly involved?”.

The litmus test should always be, is the content you are producing authentic and insightful? The likelihood is, if you are not using AI at all then you are hitting at least one of those and so you don’t have to worry about any potential impacts. If you are using AI as a final proofreader, you’ll need to be careful to ensure that your personality doesn’t get watered down by any edits. So be specific about what edits you will and won’t accept from that process.  

If however you have been using AI to generate copy and graphics from scratch, you’re likely to be affected. I’ve met many founders and teams who use AI to speed up their content production, often to try and escape the random acts of marketing cycle, and my plea to everyone is the same as it always has been. And aligned pretty well with LinkedIn’s position, “The path to insight is your own personal experience, it’s your own perspective, and I don’t think anyone is using AI to help with their personality.” 

Do not use AI to write or create for you. If you must use it, use it for tweaks and proofing. Otherwise put it aside. Do not be tempted by the short cut. In the areas of branding and content marketing, it isn’t going to serve you well. As I said in a previous article, Fast and good are not synonyms, “Do the work, embrace the friction. It’s market proof that you have done the work to learn.”