Trust Model for Virtual Assistant Based on Artificial Intelligence with Communication Between Machines

Abstract

In today’s era of Artificial Intelligence and machine learning, the whole IT infrastructure is based on AI trust and IoT (Internet of Things) which can be in terms of data or security. AI has helped human beings to live with a simple, easy, and more advanced life. To make the work easier and effective, a lot of products and services have been made using Artificial Intelligence, one of the best products is the virtual assistants which uses NLP (Natural Language Processing) to communicate and interact with humans. The human voice works as a command for the system and the task is performed accordingly. Here, trust plays a very important role as the virtual assistants know our history of tasks and our data. The virtual assistant has the capability to learn by itself and from the human behavior as it uses deep learning algorithms, so trust becomes an important topic to discuss here. As the virtual assistant has the access on our all types of data like contact, notes, our search history, locations etc. so it would also be possible for the virtual assistant to send all its data its developer’s team. the virtual assistant uses a wakeup keyword like “Alexa” used by Amazon Alexa and “Ok Google” used by Google  to wake up means that the mic is always listening to us for the wake up keyword so there may be the possibility that the assistant would be hearing and monitoring the data of the user using the camera or the mic on some particular keyword (which would be predefined) and records all those to its respective server without the permission of the user. It could also be the possibility that a keyword would be defined through which anyone can access the data of any user by just using a voice command with the keyword to the virtual assistants.