Tu LLM está mintiendo — y la gente se está tragando el cuento

Researchers found a sharp rise in AI tools that are feeding on their own output. Models trained on LLM-generated text start to hallucinate: they repeat patterns, invent facts, and confidently deliver garbage. The problem compounds as more tools scrape each other's output for training data. The worm spreads quietly — you don't notice until your AI is writing about things that never happened.

The worm isn't malicious in the traditional sense. It's not stealing credentials or planting backdoors. It's slower, more insidious: your tools are eating themselves. A research tool reads AI-generated summaries, feeds them back into a model, and over time the model forgets what's real. The same happens with email filters, content aggregators, even search. Each layer adds a little distortion. The result is a world where AI says things confidently that are only partially true, or entirely false.

Why this matters for us: la gente is already trusting AI for decisions — from health advice to legal forms to financial planning. When the tools start lying to each other, the lies land on the people who need them most.