The current president of the United States of America, Donald J. Trump, and his policies has been a controversial topic in public and social media. The social media platform Twitter offers the possibility for individuals to publicly speak their mind about political and social topics regarding these policies. Distinguishing between political posts and posts that are concerned with other matters is a crucial ability when dealing with social media. We trained a long short-term memory (LSTM) network with an implementation of word2vec -data, generated from tweets. These tweets consisted of political content as well as content unrelated to current political developments. The goal of the training process was to achieve a classifier that can distinguish between alt-right political and non-political tweets. Further, we programmed a generator that, among others, uses the previously trained classifier to create political tweets. The resulting twitter bot is active, and its tweets can be found here: https://twitter.com/TrumpFacts8/with_replies.