Voice User Interface (VUI) Design – Inappropriate Behaviour

This is a difficult subject to sum-up. On one hand it is always good fun to get a rapport going when having a conversation. By being able to do this with a ‘machine’ can make the experience feel more human and as a voice interaction it could also feel more comfortable or easy to use.

Again there have to be limits and rules on what can be said or suggested. The same rules should apply to the voice assistant as to a real person. It is not therefore OK to make racist or sexists remarks, just the same as in a personal interaction with another human being.

I am not sure how recent it is but I know that Facebook as an example have a ‘red flag’ system where inappropriate behaviour is monitored and if it continues the user can be barred from the platform. The same should be in place for a voice assistant. If a user asks inappropriate requests of a sexual or racial nature they should be warned and ultimately barred from using the skill or app if this persists.  

It is always good to have an element of humour in my opinion; it’s fun to get a funny response to a request sometimes or a back handed comment. However not everything is funny to everyone. Great care should also be taken for cultural differences and backgrounds to stop the risk of offending someone. In this case a simple ‘HI how are you today’ could be a good interaction to receive but it may not be appropriate to be asked if a ‘Health issue has cleared up’ which was discussed in a previous interaction with the app. This information is more private and it could be dangerous to broadcast this through a voice assistant. For this scenario it would be important to think about where the user is using the app and if necessary ask for further verification before discussing such matters with the user. 

As designers it’s our job to handle sensitive information correctly and to apply safeguards against harassment or abuse so that everyone can enjoy using the VUI in a friendly and safe environment.