Releasing more AI-related work (such as GPT-4 and Stable Diffusion) and projects show us new limits of automation and process optimization. These projects will automate more of today's jobs. Are we ready for that scenario? We aren't yet.
From today's perspective, it's both good and bad. It's good because humans can spend fewer resources on the same amount of work. It's bad because some people will be looking for new jobs and won't be able to find those: why pay someone for doing the routine work if a machine can do it much cheaper and faster?
To tackle the bad part, humanity should evolve: in its beliefs, systems they build, and concepts they operate.
The "evolve" part is the difficult one. To slightly initiate on that matter: humans require money to operate daily. To receive money, humans should bring value to society. What does society perceive as value? Is it a human that can do some work for others or the human itself? If we value the current operational value(=human work), a machine can bring that value for much fewer resources. If we value a human itself, we have the option to see an interesting future.