There are some fascinating early economic research papers on the impact of ChatGTP. In a nice twist, one of them even suggests that economic research is itself headed for disruption thanks for more advanced large language models https://www.nber.org/papers/w30957