In our paper we will introduce our first trial to calculate a human text input and calculate its possibility to be immoral. In the beginning we use simple SOV phrases to query the Internet and see how many immoral consequences such a phrase can cause. In the next step we apply Bentham's Felicific Calculus to calculate the degree of being negative. In the end we plan to apply our idea to a self-education system where users can decide on what they want to study about.