Blocking ads/trackers/privacy invasive APIs versus intentionally poisoning them

I see a lot of discussion on this sub about how to block certain trackers or software that tracks you and gathers you data, but I have seen few examples which try to poison the data on purpose. I have been active somewhat in the new locally ran large language models and it seems that absent data over a large data set hurts very little compared to extremely bad data injected into that data set. Obviously the larger the data …

ads apis block blocking data invasive language language models large locally lot poisoning privacy purpose software trackers try