Agentic Artificial Intelligence in Business: A Systematic Literature Review of Roles, Impacts, and Ethical Challenges

Abstract

The increasing autonomy of Artificial Intelligence (AI) systems has given rise to Agentic AI, AI agents capable of goal-driven decision-making, self-learning, and autonomous execution of tasks in dynamic environments. This evolution raises critical implications for business practices, governance, and ethical accountability. This study aims to systematically explore the roles, impacts, and ethical challenges of Agentic AI in the business domain. Using a Systematic Literature Review (SLR) methodology, 15 peer-reviewed articles published between 2020 and 2025 were selected from the ScienceDirect database, filtered for relevance in the fields of Business, Management, and Accounting. The results indicate that Agentic AI plays a transformative role in strategic decision-making, operational efficiency, and innovation, particularly through AI-powered managers, virtual agents, and autonomous service systems. However, the implementation of such systems also introduces ethical risks, including algorithmic bias, privacy concerns, and accountability gaps. The study concludes that while Agentic AI offers significant potential to reshape business processes and organizational structures, its success depends on ethical system design, human oversight, and long-term trust-building mechanisms. These findings contribute to a deeper understanding of how autonomous AI systems can be responsibly integrated into future business environments.