Good point, it’s already practically impossible to distinguish between ChatGPT o1 output and say the median substack essay. At least for someone who only has a few minutes to spare for second guessing.