The Twitch API was recently updated via Amazon's ad portal VAES for their automated ad system. Discovered via community members Tayari and Saysera, it was found that Twitch added several qualifiers to their ad targeting system including brand safety score, ban history, games played, automod enable, and manual ratings from Twitch staff. This sparked a popular Reddit thread where people surmised that Twitch was implementing a safety score for all its broadcasters.   

Of course the issue is more nuanced. In this video we discuss the Twitch ad system, how streamers make money, if Twitch can determine how much money you make, what CPMs are, and how Twitch interacts with CPMs. Is XQC a brand risk? Is TrainwrecksTV a brand risk? Are other streamers brand risks? Will a brand score effect how much money streamers get from ads? Many broadcasters believed this was the beginning of Twitch determining how much money streamers should make from advertising, but in reality this is a simple update to Twitch's ad system to better target.