Major Tech Platforms Crack Down on Low Quality AI Slop

Major technology platforms, including LinkedIn, Spotify, and YouTube, are actively purging low-quality artificial intelligence content across their services to curb automated spam and improve overall user experience. LinkedIn reported that members have used its new reporting tool over one million times since its launch on July 30, helping reduce views of AI slop by 40 percent in recent weeks. The networking platform has updated its internal classifiers to flag automated material while issuing notifications to authors whose posts draw frequent AI-related feedback from other users.

The widespread crackdown follows independent research from AI-detection firm Pangram, which revealed that 41 percent of reviewed longform LinkedIn posts were fully AI-generated, making it the most artificial among major professional networks.

According to research across major digital services between April and June, Pangram found that one in four longform posts across LinkedIn, X, Reddit, Substack, and Medium was created entirely by artificial intelligence. Other tech giants are implementing similar purges, with Spotify removing 75 million spammy tracks and duplicate uploads last year, while Google's YouTube deleted 130,000 low-quality channels over six months.

Highlighting the pervasive issue of automated content quality, content researcher Aidan Walker at the Carnegie Endowment for International Peace provided perspective on public fatigue regarding generative tools. "AI slop is just another 'festering wound,'" said Aidan Walker, content creator and researcher at the Carnegie Endowment for International Peace.

In a related discussion regarding the nuance of human-AI collaboration, columnist Mark Ritson discussed how AI detectors analyze prose fingerprints when evaluating co-authored articles. " Keeping conversations real on LinkedIn," said Vice President, LinkedIn. "Honestly, I don't know, and I'm the one who's supposed to.

I gave Mark a list of options. He wrote the thing. I tightened it against 20 years of his columns.

He rewrote my tightening, several times, and then his editor had a go. Somewhere in there, a Pangram fingerprint appeared, but a fingerprint isn't authorship. If you're asking who held the pen when it mattered, it was him.

Yes, a machine touched it. Obviously," said Claude, AI Assistant.

Platforms are continuing to adjust their detection models to curb generic, high-volume posts while aiming to preserve legitimate human writing assisted by digital tools.