New Delhi, February 16 Regulatory sandboxes are necessary to ensure that startups and other players can build things in a controlled environment within the framework, without stifling innovation, a senior official at the International Financial Services Centres Authority (IFSCA) said on Monday.
Speaking at a session at the India AI Impact Summit 2026 in the national capital, IFSCA's Chief Technology Officer Joseph Joshy also said he would look at things through a regulatory lens.
The five-day summit -- which will see a host of heads of state and government, overseas representatives, and industry players -- commenced on Monday.
"We need to ensure that there are regulatory sandboxes among regulators, so that there are guardrails... (so that) before regulations come, startups and others can build things in a controlled environment, so that you don't stifle innovation and that you encourage it within the framework," Joshy said.
Joshy is also the Chief General Manager as well as Head of FinTech & IT at IFSCA.
When there are no regulations in place for a particular area, and when entities want to innovate in that area, then regulatory sandboxes will help in charting the way forward.
According to Adani Defence and Aerospace Senior Vice President & Chief Digital Officer Rangarajan V, the country probably needs to develop its own India Context Protocols to create India-specific answers in the world of AI.
Certain aspects of Indian culture and science could be factored into them, he added.
They were participating in a session on 'India's Path to Sovereign, Frontier AI, Shaped by the needs of the Global South' at the India AI Impact Summit 2026 in the national capital.
Yotta Data Services Co-Founder MD & CEO Sunil Gupta said digital highways in terms of data centres, networks and compute capacities have to be treated as essential commodities.
"Compute infrastructure for AI is the most essential commodity...," he noted.
Meanwhile, talking to PTI after the session, Joshy said as a regulator, IFSCA will look mainly at the integrity of markets and consumer protection.
"Some of these AI models which are coming in, they might create systemic risks, and we need to be aware of it...," he said.
Regulations are not entity-based but activity-based, he said, adding that even a small startup or a big conglomerate can have a powerful model that can also have a big risk impact.
In that case, that model needs to be put within the framework. "We are not stopping anybody (from innovating)..., but ensuring to have a balance between risk and innovation," Joshy said.