The problem of epistemic circularity maintains that we cannot know that our central belief-forming practices (faculties) are reliable without vicious circularity. Ernest Sosa’s Reflective Knowledge (2009) offers a solution to this problem. Sosa argues that epistemic circularity is virtuous rather than vicious: it is not damaging. Contra Sosa, I contend that epistemic circularity is damaging. Section 1 provides an overview of Sosa’s solution. Section 2 focuses on Sosa’s reply to the Crystal-ball-gazer Objection. Section 2 also contends that epistemic circularity does not prevent us from being justified in (e.g.) perceptual beliefs, or from being justified in believing that (e.g.) sense perception is reliable. But, Sect. 3 argues that it does prevent us from being able to satisfactorily show that our central belief-forming practices (faculties) are reliable. That is, epistemic circularity prevents us from distinguishing between reliable and unreliable practices, from guiding ourselves to use reliable practices and avoid unreliable ones, and from defending reliable practices against skepticism. Hence, epistemic circularity is still damaging. The concluding section suggests that this has repercussions for Sosa’s analysis of the value of reflective knowledge.