Attacks enabled by deepfakes are an emerging problem. Much existing discussion around deepfakes relates to the concern of the increasingly realistic portrayals of real people and the potential for crime and disinformation. However, little attention is given to the fact that no technical countermeasure can defeat deepfake-enabled attacks. This is because there is not a single type of deepfake. This paper provide background on deepfakes and existing countermeasures and legislation, as well as outlines the dilemma created by deepfakes. It then uses qualitative methodology for defenders to implement when responding to deepfake-enabled attacks.