"The algorithm" and/or the people behind it noticed that if User1 watches VideoA and then we show him B-C-D, he stays on the platform for 10 minutes. BUT when

"The algorithm" and/or the people behind it noticed that if User1 watches VideoA and then we show him B-C-D, he stays on the platform for 10 minutes. BUT when we showed a User2 the videos X-Y-Z (after videoA), then User2 stayed "engaged" for 3 hours. And the new sequence was just established. The 'machine' is constantly doing A/B and other tests, and it learns, adapts, and continues. The machine just learns what people like and feeds it to them. We can't blame the machine for giving the users what they want.. can we? :)