Everyone has been saying this massive threat has been coming for years, where is it already? GPT-3.5 API costs basically nothing right now and there are increa

Everyone has been saying this massive threat has been coming for years, where is it already? GPT-3.5 API costs basically nothing right now and there are increasingly good enough local alternatives if you get banned. And before ChatGPT, GPT-3 was available to anyone with a convincing enough story and GPT-2 could be finetuned to produce decent output. I've seen https://twitter.com/xroel_ledner/with_replies https://twitter.com/xroel_ledner/with_replies and https://twitter.com/xDanial__Miller/with_replies https://twitter.com/xDanial__Miller/with_replies on Twitter which I think are almost certainly ChatGPT but they have a very inauthentic feel to them. I made something similar for Reddit for fun (https://news.ycombinator.com/item?id=36191723 https://news.ycombinator.com/item?id=36191723) but my bot accounts got most of their upvotes from reposting old stuff, not witty ChatGPT comments. I feel like good propaganda is contextual, preferably incorporates the memes of whoever its targeting, and doesn't just sound like a list of talking points. I haven't seen that yet though it is probably possible at this point. Maybe human propagandists are cheap enough for it to be not worth it?