A comparison of artificial intelligence models used for fake news detection

Abstract

This article aims to compare current state-of-the-art natural language processing models (NLP) fine-tuned for fake news detection based on a set of metrics and asses their effectiveness as a part of a disinformation management structure. The need for a development of this area comes as a response to the overwhelming and unregulated spread of fake news that represents one of the current major difficulties in today`s era. The development of AI technologies has a direct impact over the creation and spreading of misinformation and disinformation as a result of the multiple uses that technology may have. Currently, machine learning techniques are used for the development of large language models (LLM). These developments in science are also used in disinformation campaigns. Related to this matter the concept of disinformation management has arisen as a cybersecurity issue integral in the current cyber threat landscape