Explainable artificial intelligence for sustainable business performance: Integrating ESG metrics into AI adoption models

The rapid nature of the incorporation of artificial intelligence systems into the functioning of organizations has exposed a paradox that is critical in that when the algorithms of machine learning remain veiled the stakeholders cannot trust them and consequently hinder the sustainable transformation of business operations. Although there have been significant improvements in the capabilities of AI, their black-box character poses significant obstacles to their implementation in sustainability-oriented businesses, especially in the systems of environmental, social, and governance measurements and optimization. This is filled in the present research, where an integrated framework is developed and empirically supported to combine the explainable artificial intelligence principles with extensive ESG metrics to improve the performance of organizations in terms of sustainability. We used a mixed-methods design by gathering longitudinal evidence of multinationals in various sectors of the industry, spanning a specific period, using a structural equation model, hierarchical regression model, and machine learning classification algorithm as a form of analyzing the mediating mechanisms through which XAI transparency has effect on ESG performance outcomes. We realize that organizational sustainability performance scores on organizations practicing XAI-enabled systems of ESG monitoring are significantly much higher, with transparency systems that mediate the connection between AI adoption and ESG performances. Particularly, XAI implementation has been associated with a 23.7 percent increase in composite ESG scores with the environment performance responding the most. The study is theoretically useful in terms of expanding the scope of stakeholder theory and resource-based view theories to include algorithmic transparency as a strategic business resource and offers practitioners with practical models of how explainable AI technologies can be incorporated into sustainability management systems in order to generate quantifiable change in corporate environmental and social results.