Ask HN: How will social networks recognize AI-generated disinformation?

They won't. They'll become networks full of noise. To be fair, they already have. The ratio of noise to non-noise is already like 10:1. The real social communication has already moved off of these feed-based apps. It's taking place in closed group chats between people who all know each other.

What you say is interesting. A few weeks ago I created my reddit account. It's a bit complicated to get started but the discussions are so much more interesting. The fact that you have to have karma points to post is very interesting. But the problem with creating closed groups is that these people won't have access to different points of view.

Karma farming is a thing. It’ll keep out the less determined spammer but not much else.

This was already a major problem before generative AI, so not much.

It’s definitely an arms race. Why do you ask?

I read an article in the NYT talking about this topic. The naive idea is to think that we will be able to create tools with AI to manage this kind of disinformation. We all know what happened at the capitol in 2021, with an even more advanced disinformation, what could it lead to?

https://www.nytimes.com/2023/04/07/technology/ai-chatbots-go...