> And remember that academics are always required to do something genuinely new, not just "Uber for ice cream, but with AI". This hasn't been my experience. A

> And remember that academics are always required to do something genuinely new, not just "Uber for ice cream, but with AI". This hasn't been my experience. A lot of papers coming out today are quite iterative. They sound exactly like:"what if system X had feature Y". Just with more scientific jargon.