BALANCING EFFICIENCY AND ETHICS: AI AND ESG GOVERNANCE IN KERALA'S IT SECTOR

The diffusion of generative and predictive artificial intelligence (AI) across knowledge-intensive services is reshaping the organisation of work and placing new demands on corporate governance. This paper presents an integrative review of the literature at the intersection of AI-driven task transformation, stakeholder theory, and Environmental, Social, and Governance (ESG) practice, and applies it to the export-oriented information technology (IT) sector of Kerala, India. Drawing on scholarly evidence on occupational exposure, augmentation and automation, algorithmic management, and AI ethics, together with multilateral benchmark reports and park-level administrative data from Kerala’s state-backed technology parks, the review reaches three principal conclusions. First, the literature converges on a task-based view in which exposure is concentrated in routine cognitive work, while outcomes depend on whether AI is deployed to augment or to substitute labour. Second, ESG scholarship remains largely disclosure-oriented and rarely examines the workforce consequences of AI, even though the Social and Governance pillars offer a natural structure for oversight. Third, regional evidence shows export revenues outpacing direct employment growth (7.97% versus 5.50% over 2023–25; 76.6% versus 51.7% for Technopark over 2016–25). This divergence is consistent with productivity-led task automation, though it cannot establish it on its own. Drawing these strands together, the review proposes a five-stage conceptual model (Identify, Augment, Reskill, Govern, Measure) and a five-domain Governance Engine of indicators, and sets out propositions and a research agenda for empirical validation.