Small Unmanned Aerial Systems (sUAS) need to meet rigorous safety requirements when deployed in high-stress emergency response scenarios. This requires systematic testing of the hardware, software, communication networks, and human interaction points. Fuzz testing is particularly effective at finding failures in edge cases that might otherwise be missed. In Cyber-Physical Systems, such as sUAS, tests typically move from simulation to the real-world; however, tests which execute perfectly in simulation can sometimes fail dramatically when run on physical sUAS, highlighting the importance of augmenting simulations with real-world Fuzz tests. Unfortunately real-world fuzzing is not only incredibly time-consuming, but also potentially hazardous because the inevitable failures produced by effective fuzz tests can cause sUAS to physically crash or fly-away. In this talk, Cleland-Huang will draw upon her own real-world experiences of developing and validating software for sUAS applications, and will explore the ‘truths’ and ‘dares’ of Fuzzing in the Field. The truth is that in the normal course of running physical flight tests, accidental erroneous input values can cause dramatic failures and crashes, but can also be incredibly helpful at revealing points of fragility in the system where modifications are needed. However, accidental fuzzing falls far short of the systematic goals of true fuzzing, introducing the non-trivial dilemma of how to safely deploy Fuzz Testing in the field. This talk explores this ‘dare’, by exploring a systematic approach for fuzz testing physical sUAS systems, thereby empowering testers to identify real-world weaknesses in edge-cases that could have been missed in simulation. The end result is increased robustness in real-world sUAS systems.