Transport systems have a growing number of video cameras, both fixed and on-board, which are monitored visually only. Now, it is possible to infuse AI capacity to them, allowing us to classify behaviors, such as fare evasion in public transportation. With this, it is now possible to make autonomous video analysis in real time by using edge embedded technology and the processing of large volumes of data in the cloud. In addition to this, another type of proactive alerts become possible, for new applications in operations and management. The session provides real world results of edge, cloud and AI in the public transportation of the city of Valparaiso in Chile.