LOGIC OVER RHETORIC: A NYAYA INFERENTIAL ANALYSIS OF AI ETHICS

Abstract

The artificial intelligence (AI) industry has witnessed a rapid proliferation of AI ethics frameworks, in line with the widespread global adoption of AI. Such guidelines are meant to act as safeguards and typically seek to address issues like accountability, fairness, transparency, and justice. However, their practical effectiveness has been a subject of intense debate. This paper examines three such structural limitations of contemporary AI ethics frameworks: lack of enforceability, functional vagueness, and corporate capture. Using the Nyaya pancha-avayava (five-membered) inferential framework as an analytical lens, the study reformulates these critiques into structured logical propositions. The purpose is to find out if the existence of a code of conduct can realistically prevent harm. The analysis demonstrates that, in the absence of udaharana or verifiable empirical examples of non-binding functionally vague corporate-driven principles successfully restraining profit-driven harm, there can be no behavioral change. The findings provide a formal logical foundation of ethics-washing and show that until ethics frameworks satisfy the rigors of formal inference, they remain empty epistemic exercises rather than functional tools of justice.