AI Slop Is Eating Search and Training Data
Three sites generated 215,128 'best software' pages and Perplexity cited them as sources. The thread on this story is a direct demonstration of a loop that is now closing: AI generates content, AI-powered search indexes it, that content enters future training data, and the next generation of models learns from AI output rather than human knowledge.
The pattern connects to a separate thread flagging Mistral's data training opt-in problems and to general frustration that product search has become useless. One commenter put it cleanly: 'Searching for products has become impossible. If you don't already know what you are looking for, you're screwed.' That is a direct consequence of SEO-optimized AI content flooding every category.
Manipulating AI training data to make models recommend your product is already being called out as an emerging industry. This is not a future problem. It is happening now, and the defenses are not keeping up.
So what?
Founders building search-dependent discovery products are facing a structural problem, not a temporary one. If your go-to-market relies on organic search or AI-assistant recommendations, the signal quality of those channels is degrading. The builders who figure out trust signals and human-verified content layers will have a durable advantage over those who compete on volume.