The technologies known as artificial intelligence and deep learning are expanding across both the public and private sectors, and in more and more sectors of enterprise. From AI assisted medicine to AI banking systems, this growth is and will be explosive. However, these systems, while they can be very effective and efficient, are not without risks. The need to be sure that the systems, as implemented are compliant with relevant laws, governmental regulations and contractual obligations. Additionally, experience indicates that AI systems can be subject to inbuilt bias that makes the results of using the AI system suspect. This article discusses the potential problems with artificial intelligence and deep learning systems, and posits ways of mitigating potential problems while recognizing the value of these systems.