AI is good at convincing people of things, because that's what it was designed to do by companies whose only hope of ever making money is by convincing people of things. The first thing they did with it was to use it to convince people that AI is good at actual things beyond just convincing people of things, and it was and continues to be a riotous success at that, proving the concept is completely sound. Ironically this actually leads to real use cases when your own goal is to convince other people of things, but it doesn't mean that AI is actually good at anything else.
That doesn't mean it won't have consequences though or that this is just the bubble many people hope it is. The ideology wars have begun and AI is a weapon of mass destruction.