Depends on how the company ran. Fast.ai did a great study showing how some social media os designed to radacalize people. They show it as an ethics warning in c

Depends on how the company ran. Fast.ai did a great study showing how some social media os designed to radacalize people. They show it as an ethics warning in chapter 6 of their course before the chapter on generative AI. If they train models to increase engagement, said model does not care what kind of engagement it is. So facebook and youtube start the cycle of extremism before whatsapp or other apps get used to coordinate.