I think a key thing here is whether usage would increase if cost went down, for instance doctors, if the price of a medical visit decreased (ignoring the way insurance and health care works for the moment), I think people would be more likely to go to the doctor. Thus if a doctor could see more people in less time without decreasing the quality of care (which maybe AI could help with), then the doctor may be able to decrease prices and see increased usage. The increased usage would then create greater demand, which could create more jobs. (I realize doctor's not a great example for this argument, but I'm too lazy to think about a better one, lawyer might actually work better, pretend I used that in the paragraph above)

This phenomenon can be seen in how lower priced restaurants can cause people to go out more and then create more demand for restaurants. (I think this has been studied by someone somewhere)

That said this doesn't always happen, part of this has to do with the balance between lowering prices and increasing demand, etc., and part of this goes to whether or not automation can completely cover the important aspects of the job (for the typical use case) it's replacing. For example, how travel comparison and search sites cover the typical use case for a travel agent roughly (I'm guessing).