Artificial intelligence and academic integrity in the context of virtue epistemology

Abstract

The aim of this article is to examine the specificity of academic integrity in the context of the increasing use of artificial intelligence (AI) in research. Traditional epistemology does not address the specific challenges posed by this issue, so it is necessary to incorporate virtue epistemology. Proponents of this approach believe that knowledge is the result of the proper use of epistemic virtues such as openness to truth, attentiveness, and intellectual honesty. Thus, virtuous knowledge arises when a person uses epistemic virtues to arrive at the truth.

Key epistemic virtues threatened by AI are epistemic autonomy and responsibility. Autonomy allows a subject to maintain its beliefs independently, while responsibility refers to the epistemic agent's accountability for the reliability of its beliefs. In addition to virtues, we also consider epistemic vices. These characteristics of knowledge help us to formulate the concept of an epistemic agent within the framework of virtue epistemology.

Based on the concept of the epistemic agent, we examine epistemic authority, its functions, and the problems that arise in interacting with it. Do we lose epistemic autonomy when we rely on authority? Is belief based on authority rationally justified? Is a person responsible for beliefs formed through reliance on authority, especially when this leads to negative consequences?

Based on our research, we address the following questions Does AI have epistemic virtues and can it be considered an epistemic authority? The main problem is that AI is not an autonomous agent and cannot be held accountable for its information or analysis as a human authority can. AI is incapable of justifying its conclusions by epistemic virtues. Its knowledge is entirely dependent on the data provided and the methods of training, and it cannot independently assess the reliability or ethical implications of its answers. Under certain conditions, however, AI can exhibit some epistemic virtues, but this requires constant interaction with humans.

The final section of the article explores the possibility of maintaining academic integrity when interacting with AI, considering its functioning through the lens of virtue epistemology in real educational and research contexts.