Prozkoumání variant Turingova testu: Hodnocení umělé inteligence a jejího vztahu k lidské inteligenci

This thesis delves into the Turing test (TT) and its variants, examining its relevance in the ongoing discourse of artificial and human intelligence. The primary aim is to assess TT variants, evaluate their effectiveness in measuring artificial intelligence (AI) systems' human-like intelligence, and compare this with human intelligence. This involves three steps: first, analyzing different TT versions, focusing on their reliability and the complexity of AI in replicating human intelligence, including the original TT and its modifications. Second, exploring human intelligence to establish a clear benchmark for AI comparison. Third, comparing similar features of AI tools ChatGPT and Midjourney with each other and human perception using variants of TT. This thesis aims to contribute to the broader discourse on the nature of intelligence by comparing human and machine intelligence. Methodologically, the thesis is divided into theoretical research and practical analysis. The theoretical part, underpinned by a literature review, involves a conceptual analysis of human intelligence for a comparative framework. Practically, the study evaluates two AI tools, ChatGPT and Midjourney, for their demonstration of verbal and visual intelligence aspects related to the TT. These AI tools are subjected to human-operated tests, assessing their creativity and visual perception against human intelligence criteria. In conclusion, the imperfection of TT variants in assessing the presence of human behavior and qualities in AI was shown. The findings offer insights into the current state of AI and its potential trajectory, emphasizing the evolving relationship between human and machine intelligence.