> Last fall, we started using [machine learning and artificial intelligence] to extend our efforts to a range of hate groups globally, including white supremaci

> Last fall, we started using [machine learning and artificial intelligence] to extend our efforts to a range of hate groups globally, including white supremacists So... is this going to be a train wreck like YouTube's content ID where certain keywords will trigger the robots to censor my post with no human to appeal to? "Sorry, you said: '80 years ago today Hitler committed suicide and axis power was extinguished.' This post contains keywords associated with dangerous groups and individuals and has been removed."