Generative artificial intelligence is moving from experimental chatbots into the core operations of major companies, and a new study is examining whether that shift could also transform the way businesses handle environmental, social and governance responsibilities. Published in Humanities and Social Sciences Communications, the research by T. Wang, D. Lu and Y. Liu investigates the role of generative AI in enhancing corporate ESG performance, using evidence from China, one of the world’s largest and fastest-evolving technology markets. The study arrives at a moment when companies are under growing pressure to cut emissions, improve labor practices, strengthen governance and prove that their sustainability claims are backed by measurable action.
ESG performance refers to how effectively a company manages its impact on the environment, its relationships with employees and communities, and the systems it uses to make decisions and control risk. Traditionally, evaluating ESG performance has depended on extensive reporting, manual audits, scattered databases and assessments by analysts. Generative AI introduces a new technical layer to that process. Unlike conventional software, which follows narrowly defined rules, generative AI systems can process large volumes of text, financial records, satellite imagery, regulatory filings and operational data, then produce summaries, forecasts and structured recommendations. In principle, this could allow companies to detect sustainability risks faster and respond before they become costly or highly visible failures.
The Chinese business environment provides a particularly important setting for studying this relationship. China has built a vast digital economy while also expanding national commitments related to carbon reduction, green development and corporate accountability. Its companies operate across energy-intensive manufacturing, technology, transport, finance and global supply chains, creating enormous volumes of environmental and social data. At the same time, the quality and consistency of ESG disclosure can vary considerably between firms and industries. Generative AI may help bridge some of those information gaps by extracting comparable indicators from documents that were previously difficult to analyze at scale. The research therefore places AI-driven corporate transformation within a broader shift toward data-intensive sustainability management.
One of the most important mechanisms is information processing. A company attempting to measure its carbon footprint may need to combine data on electricity consumption, fuel use, logistics, raw materials and suppliers. These records often come in incompatible formats and may contain missing or contradictory entries. Generative AI can assist by classifying documents, identifying relevant variables, translating unstructured language into standardized data and flagging anomalies for human review. The same approach can be applied to workplace safety reports, employee feedback, board disclosures and compliance records. By reducing the time required to collect and organize information, AI could make ESG monitoring more continuous rather than an annual exercise performed mainly for reporting purposes.
The technology may also influence the “E” in ESG through operational optimization and green innovation. Machine-learning systems can identify patterns in energy demand, production waste and equipment performance, while generative models can help engineers explore alternative designs, materials or production processes. In a factory, an AI system might compare thousands of possible operating conditions to find combinations that reduce energy use without compromising output. In logistics, it could generate route options that limit fuel consumption and emissions. These applications do not make a company sustainable automatically, but they can give managers a faster way to test solutions and estimate trade-offs. The study’s focus is significant because it connects the rise of generative AI with the broader question of whether digital innovation can produce measurable improvements in corporate responsibility.
The social and governance dimensions are equally complex. Generative AI can analyze employee surveys, workplace incident records and public complaints to identify recurring problems that may be overlooked in conventional management systems. It can also help companies compare internal policies with labor regulations, identify potential compliance failures and prepare clearer disclosures for investors and regulators. In governance, AI-assisted analysis may reveal unusual transactions, conflicts of interest or inconsistencies between corporate statements and reported performance. Yet these benefits depend on the quality of the underlying data and the safeguards surrounding the system. A model trained on incomplete records can reproduce blind spots, while an opaque automated recommendation can make accountability more difficult rather than less.
That tension is central to the debate over AI and ESG. Generative systems are powerful because they can produce plausible language and recommendations, but plausibility is not the same as accuracy. Models may generate incorrect information, overlook minority groups, amplify historical bias or present uncertain conclusions with excessive confidence. Companies could also use AI to produce polished sustainability reports without making equivalent changes to their operations, creating a more sophisticated form of greenwashing. For that reason, AI-generated analysis must be supported by traceable data, independent verification, transparent evaluation criteria and meaningful human oversight. Corporate leaders will need to distinguish between using AI to improve performance and using it merely to improve the appearance of performance.
The Chinese evidence examined by Wang, Lu and Liu contributes to a rapidly expanding research question: whether generative AI should be understood only as a productivity tool or also as an institutional technology capable of changing corporate behavior. If AI lowers the cost of measuring environmental and social outcomes, investors may gain access to more timely information and companies may face stronger pressure to correct weaknesses. If it improves internal coordination, sustainability targets could become more closely connected to everyday decisions about procurement, product development and risk management. But if adoption is concentrated among large, well-funded firms, the technology could also widen the gap between companies with advanced digital infrastructure and those unable to afford it. The ultimate impact will depend not only on model performance, but also on regulation, organizational culture and the credibility of ESG standards.
The study appears as businesses worldwide are racing to deploy generative AI while regulators and the public demand evidence that the technology delivers benefits beyond efficiency and novelty. Its China-focused analysis offers a valuable lens on how AI, corporate governance and sustainability intersect in a major emerging technology ecosystem. The central message is both ambitious and cautionary: generative AI may help companies understand their environmental and social effects with unprecedented speed, but it cannot replace responsible leadership, reliable data or independent scrutiny. The future of ESG may be increasingly automated, yet the decisions that determine whether that future is genuinely sustainable will remain profoundly human.
Subject of Research: The role of generative artificial intelligence in improving corporate environmental, social and governance performance in China.
Article Title: The role of generative AI in enhancing corporate ESG performance: evidence from China
Article References: Wang, T., Lu, D. & Liu, Y. “The role of generative AI in enhancing corporate ESG performance: evidence from China.” Humanities and Social Sciences Communications (2026). https://doi.org/10.1057/s41599-026-08680-0
Image Credits: AI Generated
DOI: 10.1057/s41599-026-08680-0
Keywords: Generative AI, corporate ESG performance, environmental sustainability, social responsibility, corporate governance, China, artificial intelligence, green innovation, ESG disclosure
