The quality and dependability of AI analytics are proportional to the size of the Neural Networks used for any given task, and the number and complexity of the application where AI capabilities are required are only growing. At the moment the Cloud is the only platform that can deliver the kind of computational power required by the market. A new Edge platform is eagerly looked for, that can support the AI quality requirement and solve the Cloud problems in latency, security, privacy, infrastructure dependability and costs.