Interesting. And a bit creepy. I didn't know these things existed. It would be interesting to see a link to a digital influencer as an example. And I suspect the idea that they are perfect is questionable. They are subject to human error like any other program created by humans. They also cannot 'read the room' to get a sense of when something is not going well. The idea that something is more easily fixed by a bot than a human might not be right. Once something is tweeted, there are screenshots available if it's contentious enough to get a reaction. It doesn't matter whether a human or bot tweeted it. Interesting topic though.