ChatGPT: the pedagogical implications and possibilities of Artificial Intelligence

Abstract

With the latest developments in AI, however, we are facing the development of ANI or 'restricted artificial intelligence'. But what is AI and what does it consist of? How does it work? What actual potential does it have, beyond the simple function of supporting home automation, the operation of smartphones, and the response of bots (from robots) for the automatic response of companies?  Concerning the selective and decision-making processes applied to machines, there has always been a twofold possibility of reading phenomena and the activities that result from them: on the one hand, the behaviorist matrix that is very attentive to the realization of meaningful relationship processes between the first level of perception of stimuli and the responses produced by the third level, and on the other hand, cognitivism that attempts to animate the second level of input/output organization with hypotheses and cataloging processes. The central theme of artificial intelligence is bound up with the understanding of the learning mechanisms that concern man, the management of complexity through problem-solving strategies to be tackled, and the corresponding responses to be put in place. But the condition that determines the outcomes of an adequate response to the context and the demands posed by the environment, circumstances, and social demands, depends on the quality with which the questions are asked. The process described and activated at the basis of the theoretical and practical development of AI is in some respects specular to the strategies employed by subjects to solve novel problems. In this sense, there is a great deal of correspondence between biological plastic neuronal mechanisms and those of a cybernetic nature. This already-known specularity now opens up further research channels that can implement the relationship between AI and pedagogy. With ChatGPT, the process of Deep Learning evolves, which stands, on the evolutionary scale of AI, as an enhanced and evolved version of Machine Learning systems.