Blab AI
Indian Enterprises Face Security and Governance Hurdles as AI Adoption Accelerates
In a Mumbai roundtable hosted by the Economic Times, senior executives from Cisco, Reliance Enterprise Intelligence, Angel One, LIC Mutual Funds, Tata Motors, and RPG Group converged to dissect India’s growing enthusiasm for artificial intelligence and the obstacles that still keep large‑scale deployment in check.
The session opened with Cisco India and South Asia president Daisy Chittilapilly, who cited the company’s latest AI Readiness Index. According to the index, more than 80 % of organisations plan to embed AI agents in their workforce this year, yet fewer than 30 % feel prepared from a security, infrastructure or data perspective. Chittilapilly summed up the sentiment: “There is no lack of ambition, but there is significant technical debt, along with trust issues and governance challenges.”
Reliance Enterprise Intelligence chief executive Parminder Singh framed the conversation around the second phase of AI adoption. He warned that enterprises must first determine which model best suits each workflow rather than defaulting to the newest frontier model. “The winners on the podium will continue to keep changing. Don’t get too excited by a particular model,” Singh said.
Angel One’s group CEO Ambarish Kenghe highlighted the heightened risk profile in financial services, where a single AI error can have serious consequences. He identified three key risks: incorrect information generation, leakage of sensitive customer data, and unpredictable autonomous behaviour. Kenghe stressed that security must be built into AI systems from the outset, adding, “The more powerful the technology, the more careful you have to be.”
LIC Mutual Funds executive director Prashant Thakkar cautioned against chasing flashy AI for its own sake. He said the company’s priority is to equip its extensive field‑force network with relevant, personalised information. Thakkar noted that enterprises must continuously revisit resilience mechanisms and ask whether they can plug out a rogue AI system without disrupting business operations.
Tata Motors Digital.AI Labs chief information security officer Pawan Sharma called for “secure by design” principles, arguing that security should be integrated from the planning stage rather than added after deployment. He warned that AI systems no longer behave predictably and that business process owners must be involved in design discussions. “The failure of an ecosystem is a failure of imagination,” Sharma said.
Amol Deshpande, chief digital officer at the RPG Group, emphasized the need to evaluate business value, redesign processes and prepare employees before widespread AI deployment. He also warned about the growing use of synthetic data for training and predicted that enterprises will deploy “guardian agents” to monitor other AI agents, creating a good‑vs‑bad agent dynamic.
The panel agreed that India should avoid rushing into AI regulations. Instead, they called for a broad, principles‑based framework developed jointly by industry and policymakers. Singh noted that accountability remains unclear when AI systems fail, asking where responsibility lies—within the organisation, the hyperscaler, the implementer or the orchestration provider. Kenghe added that industry should self‑regulate to reduce the need for external intervention.
In summary, Indian enterprises are eager to adopt AI but face significant technical debt, governance gaps and security bottlenecks. While many organisations plan to integrate AI agents into their workforce, less than a third feel ready from a security and infrastructure standpoint. The discussion underscored the importance of workflow‑specific model selection, secure‑by‑design principles, continuous resilience testing, and a collaborative regulatory approach. The path forward will require coordinated efforts across technology, governance and policy to ensure trustworthy AI deployment at scale.
The session opened with Cisco India and South Asia president Daisy Chittilapilly, who cited the company’s latest AI Readiness Index. According to the index, more than 80 % of organisations plan to embed AI agents in their workforce this year, yet fewer than 30 % feel prepared from a security, infrastructure or data perspective. Chittilapilly summed up the sentiment: “There is no lack of ambition, but there is significant technical debt, along with trust issues and governance challenges.”
Reliance Enterprise Intelligence chief executive Parminder Singh framed the conversation around the second phase of AI adoption. He warned that enterprises must first determine which model best suits each workflow rather than defaulting to the newest frontier model. “The winners on the podium will continue to keep changing. Don’t get too excited by a particular model,” Singh said.
Angel One’s group CEO Ambarish Kenghe highlighted the heightened risk profile in financial services, where a single AI error can have serious consequences. He identified three key risks: incorrect information generation, leakage of sensitive customer data, and unpredictable autonomous behaviour. Kenghe stressed that security must be built into AI systems from the outset, adding, “The more powerful the technology, the more careful you have to be.”
LIC Mutual Funds executive director Prashant Thakkar cautioned against chasing flashy AI for its own sake. He said the company’s priority is to equip its extensive field‑force network with relevant, personalised information. Thakkar noted that enterprises must continuously revisit resilience mechanisms and ask whether they can plug out a rogue AI system without disrupting business operations.
Tata Motors Digital.AI Labs chief information security officer Pawan Sharma called for “secure by design” principles, arguing that security should be integrated from the planning stage rather than added after deployment. He warned that AI systems no longer behave predictably and that business process owners must be involved in design discussions. “The failure of an ecosystem is a failure of imagination,” Sharma said.
Amol Deshpande, chief digital officer at the RPG Group, emphasized the need to evaluate business value, redesign processes and prepare employees before widespread AI deployment. He also warned about the growing use of synthetic data for training and predicted that enterprises will deploy “guardian agents” to monitor other AI agents, creating a good‑vs‑bad agent dynamic.
The panel agreed that India should avoid rushing into AI regulations. Instead, they called for a broad, principles‑based framework developed jointly by industry and policymakers. Singh noted that accountability remains unclear when AI systems fail, asking where responsibility lies—within the organisation, the hyperscaler, the implementer or the orchestration provider. Kenghe added that industry should self‑regulate to reduce the need for external intervention.
In summary, Indian enterprises are eager to adopt AI but face significant technical debt, governance gaps and security bottlenecks. While many organisations plan to integrate AI agents into their workforce, less than a third feel ready from a security and infrastructure standpoint. The discussion underscored the importance of workflow‑specific model selection, secure‑by‑design principles, continuous resilience testing, and a collaborative regulatory approach. The path forward will require coordinated efforts across technology, governance and policy to ensure trustworthy AI deployment at scale.
