Vendors are emphatic that content metadata holds myriad value for broadcasters, increasing video’s findability and searchability, tracking rights and protecting against deepfakes and using AI to make the entire process easier. Above: An example of using machine learning to identify objects in video with correlating time markers. In this way, users can search across video with increasingly sophisticated queries such as “2 Persons + 7 Cars” to show all matches across an entire content library.