By intent, or a by-product of it's propaganda machine ?
I go towards the latter. Because if they flood certain areas of the web with their misinformation and along come the scrapers to feed into their LLM trainers. I have seen AI posts on here where click the little blue number reference, and it's reddit.
And that leads into model collapse.
Model Collapse: What Happens When AI Trains on AI-Generated Data (2026) | AI Safety Directory (aisecurityandsafety.org)
As more and more AI slop is produced, more and more of it is fed into the system. So it degrades.
When China same out with their AI LLM, Deepseek, I did wonder if they limited it's learning data to what was available behind the Great Firewall. And it appears not. Because there was human programmed intervention for certain "sensitive" questions. In fact, I recall at the time there was accusations it used other peoples data sets.
On this, was the copyright thing ever sorted ? I don't think it has been.
I suppose it's all a bit like Trump University. I think if most posters here saw that at the top of a CV, we would form opinions. But do AI scrapers ?