In the near future, uninhabited aircraft system (UAS) manufacturers may begin submitting their systems to the FAA's certification process. In anticipation of this possibility, there is a need to determine the human-machine interface (HMI) certification requirements for UAS and whether those requirements are specified in existing FAA regulations. UAS piloting involves cognitive work, such as situation assessment, problem detection, and decision-making, and this cognitive work relies heavily on the design of the HMI. Accordingly, we used a method for assessing complex cognitive work, cognitive task analysis (CTA), to identify HMI design risks to safe UAS control. Identified risks were compared against existing FAA regulations and guidance. Findings suggest that new regulations and guidance may be needed to address new risks posed by UAS datalink latency and disruption, handoffs between ground control stations, new threats to cockpit sterility, and new information display requirements. These findings are compared to the findings of a traditional task analysis conducted for a similar purpose. The comparison supports the validity of the CTA results and presents a more complete gap analysis than is achieved by either analysis alone.