AI plays a vital role in understanding patterns which could result in identifying business insights. However, doing this at the level of a cloud is not enough. In the new scenario, where billions of divices are connected, the computing must be done and data processed on the device itself. If the data is on the cloud and is to be accessed from there, there will be a delayed response. This delay is called latency. It could be in milliseconds, but even that is detrimental in some cases, say a car being driven on the streets or a remote surgery being performed.

Edge computing assisted by AI plays a vital role here. It processes information on the device itself or on a platform closest to the application. Effectively it is at the edge of the networks. Cloud computing is remote and is deep inside the network.

IoT is expanding market. Edge AI computing enables billions of such devices to process information and manage outcomes. It reduces latency. With the advent of 5G, the connectivity of the devices will go up further, and the need for edge AI computing will be more too.