I know MediaBiasFactCheck is not a be-all-end-all to truth/bias in media, but I find it to be a useful resource.
It makes sense to downvote it in posts that have great discussion -- let the content rise up so people can have discussions with humans, sure.
But sometimes I see it getting downvoted when it's the only comment there. Which does nothing, unless a reader has rules that automatically hide downvoted comments (but a reader would be able to expand the comment anyways...so really no difference).
What's the point of downvoting? My only guess is that there's people who are salty about something it said about some source they like. Yet I don't see anyone providing an alternative to MediaBiasFactCheck...
I think you're missing the key part of the problem. It isn't the AI that's the issue.
The problem is that he was being paid for how many listeners his AI songs got. But he used bots to "listen" to the songs. Nobody actually listened to his AI music.
The flaw in the system was that they couldn't detect his bots. (And the bots are not AI)