In future mobility services using highly automated vehicles on the road (HAV; SAE 4), the lack of a human driver in the vehicle means that users will have no one to communicate directly with in case of problematic situations. Especially, if the HAV has to perform a minimal risk maneuver (MRM), i.e., in case a HAV’s capabilities are exceeded, users might experience insecurities. One possible countermeasure could be an internal Human-Machine Interface (iHMI) onboard the shuttle providing information to shuttle users. However, it is still unclear how information should be communicated to users to ensure good user experience during MRMs. This experimental study in virtual reality (VR) uses the media-richness theory to address this issue by providing a multi-layered iHMI consisting of a visual representation of upcoming steps during MRMs accompanied by an avatar surrogating the media richness of face-to-face communication with a shuttle driver. To test the influence of different iHMIs on user experience participants were able to experience three VR-simulated MRMs in an automated shuttle. Using a block-design, participants experienced three versions of the iHMI in each situation. They were either shown solely the visual interface with the problem-solving steps, the visual interface including the avatar’s voice, or the visual interface in addition to the avatar including its voice. After each situation, participants were asked to answer several questionnaires regarding their experience. Results provide insights into how the media-richness of iHMIs affects HAV passenger’s user experience during MRMs and, therefore, facilitate user adoption of these future mobility services.