Tried my first experiment with OpenAI's DeepResearch, asking it to evaluate the "poptimism" debate from a new framework. Then I fed the output into Claude to make the writing better. I think it's high quality, yet riven with hallucinations (made up quotes)