LogFAQs > #964844103

Honestly, I have been thinking of this and I do think there should be a dedicated group committed to identifying how dogma spreads and how to prevent it in a scientific analysis sort of way, which could then go and put forward open source industry standard recommendation algorithms to social media groups. That's more open than "the government does it", though.

My vision of that would basically entail "what amount of random content needs to be injected into social media recommendations to mathematically prevent echo chambers while allowing as much customer retention as possible". Tl Dr an anti-algorithm algorithm.

In case I wasn't clear I think the root of the problem is that social media algorithms create feedback loops and echo chambers that give rise to more reality-detached ideas because their main driving force is customer retention. They use machine learning to hone it to a point, they don't even know what it's doing. For all they know the AI could be intentionally radicalizing people.