Since there's no objective definition of "hate content", this is akin to them patting themselves on the back for outright censorship based on their own arbitrary definitions.
They clearly outline their definition of "hate content" in their post about it. [0] They go over the methodology and dig a bit into the results. It looks like a good start. [0] - https://www.reddit.com/r/redditsecurity/comments/idclo1/unde...
Am I missing something here? I read the post and it's not clearly defined at all. All they did is mention was:
>To generate and categorize the list of keywords, we used a wide variety of resources and AutoModerator* rules from large subreddits that deal with abuse regularly.