I wonder when we’ll see the first form of AI governance. Instead of large government, have an AI agent optimize for desired outcomes.

Would be interesting to see if anyone is already testing this.