The images from VK included in the author's original post are somewhat disturbing. Nothing in this incident is inconsistent with an automated process flagging

The images from VK included in the author's original post are somewhat disturbing. Nothing in this incident is inconsistent with an automated process flagging his article, and a underpaid/overworked reviewer seeing the images without taking time to digest the article in full and concluding "yup, looks like hate speech". Aside: I helped moderate a (poorly-run) veganism Facebook group for a few months. If I banned everyone who posted pictures of eviscerated animals, I'd have banned half the legitimate members. (There's a certain type of activist that thrives on eliciting visceral reactions from those who already agree with them. I don't get it.) So I can understand how Google's review process could get tripped up by stuff like this.