At BW Privacy World’s inaugural panel, industry leaders urged India to balance innovation with accountability as AI reshapes data protection and compliance norms
As artificial intelligence (AI) continues to transform data ecosystems globally, experts at the BW Privacy World panel on “The AI–Privacy Nexus: Governing Data in the Age of Intelligent Machines” emphasised the urgent need for India to align its privacy governance frameworks with global standards while nurturing indigenous models for responsible AI adoption.
Moderated by Kriti Sharma, Regulatory Legal, Compliance & Ethics Head for India and Southeast Asia at Dun & Bradstreet Information Services India, the discussion explored how AI and machine learning are reshaping compliance, data protection roles, and ethical accountability.
Bridging India’s DPDP Act With Global Standards
Opening the discussion, Ibrahim Khatri, Founder and CEO of Privezi Solutions, said that India’s Digital Personal Data Protection (DPDP) Act offers a distinct framework that acknowledges the evolving risks of intelligent systems.
“The DPDP Act talks about anything that harms the rights and freedoms of data principals it’s broader than a simple privacy violation,” Khatri noted, explaining how significant data fiduciaries handling large volumes of sensitive data would play a crucial role in AI-driven risk assessments.
Khatri highlighted use cases from the insurance sector, where AI models influence underwriting and risk decisions. “Understanding model risk, data lineage, and how AI influences such outcomes is essential to create a trustworthy AI ecosystem,” he added.
Defining Responsibility In AI Deployments
Addressing the question of accountability in AI-led decisions, Anand Kumar Kasturi, Executive Vice President and Data Protection Officer at Kotak Mahindra Bank, argued that defining ownership upfront is key.
“Liability doesn’t stick unless responsibility is defined,” Kasturi said. “In any AI system, there’s a deployer, a developer, and a victim, and we need a clear responsibility matrix for each.”
He cautioned against deploying AI without robust guardrails: “AI is like building a Ferrari, brakes cannot be an afterthought. You can’t stop midway and retrofit ethics into it.” Kasturi also noted that organisations must establish pre-deployment policies and post-implementation evaluation mechanisms, adding that “AI never unlearns”, making rollback strategies equally critical.
Legal, Cultural Challenges In AI Regulation
Advocate (Dr) Prashant Mali, Founder of Cyber Law Consulting (Advocates & Attorneys), highlighted that India is already moving toward defining AI regulation through its IT Digital Media Ethics Code and emerging AI labelling norms.
“The government has started defining what synthetic data is and how it should be labelled with metadata,” Mali said. “This will have both legal and cost implications for corporates.”
He warned that over-reliance on synthetic data could distort cultural and cognitive realities. “AI will start governing our thoughts if unchecked. The bigger issue is — are we even prepared for AI regulations? Most organisations don’t have the budget or visibility to manage their data today,” Mali cautioned.
Building Trust, Accountability In AI Systems
From a market infrastructure perspective, Sushil Ostwal, Chief Data Officer at BSE India, emphasised that responsible AI adoption must ensure both investor safety and technological innovation.
“At BSE, we are experimenting with AI but ensuring every step is aligned with regulatory safeguards,” Ostwal said. “Continuous monitoring of vulnerabilities, model drift, and data accuracy is critical to ensure no incorrect decision affects customers.”
He added that human oversight remains central: “In our case, humans and machines work together. We benchmark model accuracy and investigate immediately if results deviate from expectations.”
Moderator Kriti Sharma concluded that while India must harmonise with frameworks like GDPR and CCPA, it also has the potential to define its own model for ethical AI.
“We are at a stage where due diligence alone is not enough. There needs to be an AI-specific checklist aligned with global benchmarks but suited to India’s governance environment,” Sharma said, calling for convergence between compliance, innovation, and transparency.
As the discussion highlighted, balancing technological advancement with ethical responsibility remains central to shaping India’s AI future, where privacy, accountability, and innovation must evolve together.
