In rough order of priority (highest to lowest):
a) I do not believe that it produces useful-quality results for almost any of my use cases (Even if other concerns were addressed, I don't want to generate buggy code I then have to debug).
b) I believe many if not necessarily all generative AI models were trained unethically on copyrighted data
c) I am worried about the legal & copyright implications of using the tool to produce work products for my employer
d) I am concerned about the societal implications of the tool's existence, and how it's producing sub-human-scale work for (nearly) free, thereby eroding the market for a lot of human-produced output.
e) I dislike the amplified volume of low-quality content online over the last few months.
f) I'm not a fan of OpenAI as a business or a potential monopoly