First Impessions on Using Ai Powered Chatbots, Tools and Search Engines: Chatgpt, Perplexity and Other - Possibilities and Usage Problems


Aleksandra Fostikov




In a world that has already stepped into the Age of Imagination, an era that directly follows the Digital Age, the rise of the use of new technologies has led to the emergence of not only augmented and virtual reality and the accelerated development of robotics, but also to a leap in the development of artificial intelligence. Its outlines are clearly visible in the latest model of ChatGPT, a product of the company OpenAI that became available to the public in open access on November 30th, 2022. It is believed that by December 4th, 2022, it already had one million users. There were already over 250 scientific articles on the topic of this model by February 12th, most of them published in the form of preprints on portals such as Academia.edu, ResearchGate, and even Arxiv, and it found its place in eminent magazines such as the Economist and Forbes, among others. The name of this AI bot appeared in thousands of daily newspaper headlines, including in the domestic Serbian daily newspaper Politika (Политика), and continuously appears in everyday reports from various fields that are also published by electronic daily portals. In addition, the news about it gained significant number of followers as the AI itself on the social and business networks such as LinkedIn. Given that this is the beginning of a new era both for numerous sciences as well as for the civilization, this article is divided in three parts. According to said, the first part pays attention to early impressions of the use of this model and its possibilities, the second analysis its application so far and its impacts, while the third part reviews its possible applications in digital humanities, archivists and museology. Additionally, as the first search engines based on the use of this model appeared in the meantime, attention is also drawn to their capabilities. This paper is written based on the material and articles published so far, as well as on testing of this model and its related search engines.