By Viswanath Ramaswamy, Vice President, Technology, IBM India & South Asia
For all the enthusiasm around artificial intelligence, one truth is now impossible to ignore, AI will only be as valuable as it is trusted. Across boardrooms and policy circles, the conversation has shifted from what AI can do to what it should do. The difference between these questions may well determine whether AI becomes India’s growth engine or its governance challenge. Responsible AI, once dismissed as a moral luxury, is now recognized as a serious business advantage.
The ROI of Doing the Right Thing
Recent research by the IBM Institute for Business Value reveals that Indian enterprises investing early in AI ethics are seeing measurable business returns: 74% report an increase in customer trust, 55% say their brand reputation has strengthened, and 53% have mitigated reputational risks.
These are not intangible gains. They can directly influence customer retention, investor confidence, and the pace of AI adoption. In fact, globally, companies that fall within the top quartile of AI ethics investment have seen 34% higher operating profit from AI in 2023 than those in the lowest quartile.
What’s more striking is the trajectory, Indian organizations expect to double their AI ethics budgets, from 5.7% of total AI spend in 2023 to nearly 11% by 2026. This acceleration reflects a growing realization that ethics does not slow innovation, it enables it.
From Compliance to Competitiveness
The era of minimal compliance is over. As India advances its Digital Personal Data Protection (DPDP) framework and prepares its national AI governance policy, enterprises are being pushed to go beyond regulatory readiness. AI ethics is no longer about checking a box, it is about building systems that people, regulators, and markets can trust.
Yet challenges persist. 66% percent of Indian executives admit that building trust in AI is a major challenge within their organizations, and 53% say that ethics, bias, and explainability remain key barriers to adoption. This suggests that many enterprises still treat responsible AI as a governance add-on rather than a design principle. If the last few years were about experimenting and scaling AI models, the next few will be about scaling AI trust.
Trust as an Operating System
AI ethics is not a set of PowerPoint principles or training modules. It is a business operating system that runs beneath every algorithm and decision model. Organizations that operationalize AI ethics through bias audits, AI risk assessments, transparent model documentation, and employee training are seeing visible impact, higher adoption, fewer customer complaints, and better product quality.
In India, enterprises that invested in responsible AI have seen a 22% improvement in customer metrics and a 20% reduction in AI-related incidents in the past year alone. When ethics is embedded into the core of the AI lifecycle, from data collection to deployment, it creates a self-reinforcing loop of trust. Employees are more willing to use AI tools, customers are more comfortable engaging with them, and regulators are more confident approving them.
Agentic AI and the Governance Frontier
The emergence of agentic AI, systems capable of taking autonomous actions, adds urgency to the ethics debate. 70% of Indian executives agree that these systems will need to rethink of their AI ethics frameworks significantly.
This is a wake-up call for India Inc. The promise of intelligent agents that can write code, negotiate transactions, or take procurement decisions also comes with an elevated risk of unintended consequences. Without robust governance, even a small lapse in explainability or data integrity can cascade into systemic failures.
Embedding Ethics into Strategy
For Indian enterprises, this means re-imagining AI ethics as a growth strategy, not a cost center. The organizations that will lead are those that: a) treat trust as a business metric, tracked alongside ROI and productivity, b) embed cross-functional accountability, bringing together data, design, risk, and security teams under a unified AI governance framework, c) invest in human-centered design, ensuring AI decisions remain explainable and equitable.
Enterprises that recognize this early will reap what I call the “trust dividend”, higher adoption, stronger stakeholder confidence, and protection from reputational shocks. Those that delay will find themselves building AI systems that no one fully trusts, and therefore, no one fully uses.
As India crafts its national AI policy, the message for enterprises is clear – Don’t wait for regulation to tell you what responsible AI looks like. Build it into your business DNA now.
