Whalesbook
Indian Enterprises Elevate Responsible AI to Strategic Priority
Responsible Artificial Intelligence (RAI) has transitioned from an ethical consideration to a core business objective for Indian enterprises, according to Nasscom's 'State of Responsible AI in India 2025' report. The study, conducted between October and November 2025 among 574 senior executives, indicates a strong correlation between an organization's AI capabilities and its maturity in responsible governance practices. Nearly 60% of firms confident in scaling AI have established robust RAI frameworks, underscoring its link to trust, governance, and long-term value generation. This marks a clear year-over-year advancement since 2023, with 30% of businesses now reporting mature RAI practices and 45% actively implementing formal frameworks.
Sectoral and Enterprise Maturity Insights
Large enterprises continue to lead in RAI maturity, with 46% reporting advanced frameworks. In contrast, 20% of small and medium enterprises (SMEs) and 16% of startups demonstrate similar maturity levels, though willingness to adopt responsible AI norms is growing across smaller firms. From an industry perspective, the Banking, Financial Services and Insurance (BFSI) sector leads with 35% maturity, followed by Technology, Media, and Telecommunications (TMT) at 31%, and healthcare at 18%. Nearly half of businesses within these sectors are actively enhancing their RAI frameworks. Sangeeta Gupta, Senior Vice President and Chief Strategy Officer at Nasscom, emphasized that responsible AI is fundamental to trust and accountability, especially as AI becomes integrated into vital sectors like finance and healthcare. She advocated for embedding responsibility throughout the AI lifecycle, moving beyond mere compliance.
Workforce Development and Evolving Accountability
A significant focus for organizations is workforce enablement, with close to 90% investing in AI sensitisation and training programs. Companies expressed the strongest confidence in their ability to meet data protection obligations. Accountability structures are also developing, with 48% of organizations assigning AI governance responsibilities to the C-suite or board, while 26% delegate it to departmental heads. AI ethics boards and committees are increasingly being established, particularly among mature organizations, with 65% having adopted such bodies.
Persistent Hurdles and Future Risks
Despite the advancements, significant challenges continue to impede effective RAI implementation. The most commonly cited AI risks include hallucinations (56%), privacy violations (36%), lack of explainability (35%), and unintended bias or discrimination (29%). Primary obstacles to successful RAI include a lack of high-quality data (43%), regulatory uncertainty (20%), and a shortage of skilled personnel (15%). SMEs, in particular, highlight high implementation costs as a critical constraint. As AI systems become more autonomous, businesses with higher RAI maturity feel better equipped to address emerging technologies like Agentic AI. However, industry leaders caution that current frameworks will likely require substantial updates to manage the unique risks posed by these advanced autonomous systems. The quality and accessibility of data, particularly for Indian languages, remain a key impediment, despite India's vast data generation capabilities.