Raw LLM Responses

These bots can be trained for maximum effectiveness. The people who do this can try multiple experiments to see what works best, and train an AI algorithm to do exactly that - length of posts, subject matter, depth of vocabulary, how many posts to make off-topic, and on and on. You can make a hundred variables. You can even just do a training reinforcement, and get the most effective campaign with an algorithm nobody understands or directly wrote. They’ll just get better at it. Much better than actual people could be. So you have to wonder, what kind of people do this? Dishonest people, that’s who. They’ll win because being dishonest is more effective than being honest.