The subject of this paper is automatic text summarization in the context of organization, presentation and search of information. Automatic text summarization is applicable in many fields; however, this paper examines the role of text summarization in the domain of informational sciences. This paper presents different outlooks on text summarization and the primary methods of performing said task. Apart from two main, extract and abstract methods, a hybrid method and present-day metrics are also mentioned. Furthermore, the paper deals with shortcomings of methods in text summarization and the process of the summarization itself, which are redundancy, financial investments, double meanings, synonyms, cohesion and coherence. Nevertheless, automatic text summarization is helpful in processing text documents, especially with today's exponential increase of such information. The final goal of the text summarization field is to create high-quality summaries similar to those produced by humans, but without the need to monitor the summarization process. The evaluation of text summarization is emphasized, as well as the possible advancements of summaries created by tools for text summarization during the evolution of NLP and artificial intelligence in general.