Harry Collins interprets Hubert Dreyfus's philosophy of embodiment as a criticism of all possible forms of artificial intelligence. I argue that this characterization is inaccurate and predicated upon a misunderstanding of the relevance of phenomenology for empirical scientific research.