Today, we are exposed to much information, ranging from the media, daily news, entertainment programs, and even advertisements to widespread social networks with a powerful content dissemination scope. This leads to a bombardment of information that can be overwhelming and makes it challenging to distinguish truthful information from other types of information that have been manipulated to influence the thinking and actions of consumer users. In addition to bait links, also known as “clickbait”, this uses flashy headlines or extravagant images that are out of touch with reality to increase web traffic, diverting attention from more culturally valuable content and encouraging the consumption of superficial and unhelpful information. To combat this type of situation, the Moral Sentiment Analysis in Textual Data (AMOR) project arises from the hand of the Intelligent Systems Group (GSI) belonging to the Technical University of Madrid (UPM). This project aims to develop critical thinking and the ability to manage emotions when reading media and social networks and fighting against disinformation, hate speech, and bait links. This would help assess the credibility of the information sources consulted and develop an opinion not based on hate speech or influential news. This project is divided into different general objectives. Our purpose focuses on using social robots that interact with humans and being able to store all these interactions in what is called an ethical black box so that we can represent the information using Knowledge Graphs (KGs) and use different tools to access this information and audit it in the event of any behavior that is out of the ordinary. As mentioned earlier, this system associated with the problem allows us to keep a record when dealing with users with different personalities, cultures, or beliefs to detect and analyze the behavior of these users, and the social robots can adapt to the type of person they are dealing with.