There is a lot of cope about what will happen to humans once we have artificial intelligence that surpasses humans at every task.
Some argue that taste or agency will still matter. Both are still intellectually bound though: agency matters insofar as it's applied to working on the right things, and knowing what things are right to work on is downstream of intelligence. Similarly for taste, which is today thought of as an intuitive process. The reason it is intuitive is because the processes involved in what pleases people aesthetically are dynamic and too complex for a human to model at a purely intellectual level. That is not the case for a sufficiently advanced artificial intelligence, which can consume and parse exponentially more data.
The reality is that once artificial intelligence surpasses all humans, we won't be useful anymore. That is not to say that we won't find ways to stratify society. But it won't be based on actual contribution to civilization, as it is today, given that the mental power of a human will be negligible compared to what a machine contributes. My best guess is that we will take part in purely competitive games, just like we do with chess (in which machines already dominate us).
Personally, I find this a terrifying thought, because a future of endless abundance (which is the best-case scenario) feels to me like living in a zoo. Sure, all our needs are met on a physical level, but our desire to be useful seems deeply ingrained and one will no longer be able to fulfill this desire.