eXplainable AI for the detection of faux on social media

In some areas, black-box technological solutions work very well, as understanding the internal processes of an Artificial Intelligence (AI) model may not always be necessary. However, it has been demonstrated that Artificial Intelligence models may base their analyses on wrong assumptions, which results in bias. In areas where technology deals with sensitive data, such as medicine, law, ethics, politics or employment, such bias could prejudice the welfare of individuals, groups and organizations. In those cases, the use of eXplainable Artificial Intelligence (XAI) is crucial, as its white-box approach means that human experts can interpret the internal workings of the model and thus, ensure that responsible decisions are made. In other words, XAI makes it possible for humans to rely on technology in scenarios where opaque technologies could not be used