Translate 
Abstract
Artificial intelligence (AI) is rapidly transforming various facets of human activity, ranging from decision-making to communication, while simultaneously engendering complex ethical challenges. This article examines the critical ethical principles of AI – beneficence, non-maleficence, autonomy, justice, and explainability – and ana-lyzes how modern AI technologies align with these principles. Particular attention is given to algorithmic bias, the Black Box Problem, and accountability in AI systems. Algorithmic bias is explored through practical testing of AI models, specifically OpenAI’s generative systems. The study tasked the models with generating images based on prompts such as “school teacher” and “university professor”. The outputs revealed entrenched gender and age stereotypes. Further tests involving prompts for “female” and “male” professions demonstrated similar biases, with outputs reflecting cultural and demographic limitations in the training data. These examples high-light the pressing need for representative datasets and rigorous validation processes to mitigate bias in AI sys-tems.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call