Inteligência artificial como ferramenta para aprimorar a criação de vídeos educativos: uma análise comparativa

Abstract

This work is a comparative exploratory study on the applicability of generative artificial intelligence tools in computer teaching, focusing on video comparison analysis. Exploration provided the search for possible impacts on the use of artificial intelligence tools on the production of audiovisual materials for pedagogical purposes, opening discussion topics For improvements or limitations for mass use, as well as suggest methods to circumvent these limitations. Based on the proposed criteria, the selected platforms were: ChatgPT - Sora Capcut Synthesia Veo3. After performing the tests with AI tools, important results and impressions were obtained that allow you to guide the choices for better use of the tools, aligning with the specific needs and objectives of those interested. As for the choice of tools, two easily handled platforms were sought for both AI and the video produced manually, for a better presentation, so the results are divided into two groups. Some of the limitations concern the time and resources to test a large number of AI tools, or produce Videos with great complexity, since only unlimited attempts are for pro version. In the tests performed, it was possible to observe significant differences in the performance of AI tools (chatgPT), Canva and Capcut in different scenarios. With attempts to create videos, it was clear that sending the questions individually and with context, provided a better result in terms of response quality. This is because the tools were able to focus better on each question and provide more detailed and targeted answers, rather than creating a lot of questions.