The Importance of Utility Maximization in the AI Economy

There are not that many systems doing something like utility maximization in the new AI economy. It is unclear whether a particular qualitatively identified force for goal directedness will cause disaster within a particular time. In coherent AI's are never observed making themselves more coherent, and training has never produced an agent unexpectedly. There are lots of vaguely-agientic things, but they don't pose much of a problem. Some amount of AI divergence from your own values is probably broadly fine, that is, not worse than what you should otherwise expect without AI.