National Technology Day 2025: Agentic AI – India's Next Tech Frontier? For Agentic AI to truly thrive, experts say that businesses must embrace more than just technological upgrades

"Agentic AI represents a fundamental advancement in artificial intelligence. It means creating AI systems with true agency—ones that can perceive, reason, plan, and take action independently," said Jensen Huang, CEO of NVIDIA, during his keynote at GTC 2025. Echoing this vision, financial commentator Raoul Pal noted in a podcast, "Agentic AI will independently build entire businesses, revolutionising competition and innovation."

But are Indian businesses ready to embrace this era of intelligent autonomy? Deloitte's fourth wave of the State of GenAIreport (India perspective) found that over 80 per cent of Indian organisations are exploring the development of autonomous agents, suggesting that the business landscape is leaning strongly toward Agentic AI. However, while optimism is widespread, the journey from exploration to execution remains complex.

Apurv Agrawal, Co-founder and CEO of SquadStack, believes there are foundational realities that businesses must accept. He explained that building Agentic AI is not simply about deploying APIs or refining large language models. "It's about re-architecting how your business thinks. You're designing systems that don't just respond, but act autonomously with context, judgement, and accountability." He noted that for enterprises starting from scratch, the cost could range from INR 50 crore to INR 200 crore over three to five years. And even then, deployment success is not guaranteed unless systems, data loops, and change management are bulletproof.

According to Agrawal, while a large percentage of Indian leaders express intent to adopt Agentic AI, that intent doesn't automatically translate into capability. "We're still in the early innings. India will see pockets of meaningful adoption within 12 to 18 months, but broad-scale adoption will take three to five years," he said.

Narendra Sen, Founder & CEO of RackBank and NeevCloud, agreed that a paradigm shift is essential. "Indian businesses must evolve from task automation to intelligent, context-aware decision-making. AI needs to be treated not just as a tool but as a collaborative partner."

Early results industry leaders are seeing

Suryanarayanan Ramamurthy, Head of Data Science, Contentstack, shared that infusing Agentic AI into their platform led to a noticeable uptick in customer take rates and enabled scalable personalisation. "We're piloting workflows that help automate complex tasks and support faster decision-making. The impact is clear even at this early stage."

Nishant Rathi, Founder and Director, NeoSOFT, pointed to improved resolution times and internal efficiencies. "In customer service, autonomous agents reduced average resolution time by nearly 40 per cent. They're also generating test cases, recommending code refactors, and even streamlining onboarding through knowledge agents."

Beerud Sheth, CEO of Gupshup, said companies using their AI agents saw productivity improvements of at least 40 per cent. "We've seen lead conversion rates double or triple, especially in sectors like real estate, e-commerce, and financial services."

Ganesh Gopalan, Co-founder and CEO of Gnani.ai, echoed similar outcomes. "Our deployments reduced customer support costs by up to 80 per cent while significantly improving CSAT scores. These agents don't just replace humans, they augment them by automating routine tasks."

Kanakalata Narayanan, Vice President of Engineering, Ascendion, noted the significance of evolving AI standards. "With frameworks like the Multi-Agent Collaboration Protocol, we now have digital specialists that reason, collaborate, and only escalate to humans when required."

From hype to implementation

For Agentic AI to truly thrive, experts say that businesses must embrace more than just technological upgrades. Moumita Sarker, Partner, Deloitte India, emphasised the need for democratising AI use across teams and building libraries of reusable agents. "It's not just about using agents but allowing them to own iterative, non-standard tasks."

According to Narayanan, leadership must treat Agentic AI as a strategic investment. "This isn't just about upgrading software; it's about transforming the workforce. We need governance, experimentation budgets, and rapid prototyping."

Gopalan added that businesses often stop at automation but fail to embrace AI as an evolving collaborator. "Success lies in building robust data infrastructure, fostering a culture of continuous learning, and enabling human-AI synergy."

Vara Kumar Namburu, Co-founder of Whatfix, believes the focus must return to people. "AI must simplify workflows, not create friction. Our philosophy of 'Userisation' puts the user first. Technology must adapt to people, not the other way around."

Barriers in scaling

Despite the momentum, infrastructure and workforce readiness remain hurdles.

"Agentic AI is compute-hungry," said Sen. "It requires high-frequency inferencing, orchestration, and integrations. At RackBank, we're building India's first full-stack AI infra platform to address this."

Sheth added that legacy systems continue to be a bottleneck. "Many enterprises can't scale AI because their systems don't support modular integration. That's where our API-first approach helps bridge the gap."

Mahesh Parab, Partner - Agentic Automation from PwC highlighted another challenge: hallucination control and agent lifecycle management. "We need frameworks that make agents behave responsibly, are auditable, and align with business KPIs—just like human employees."

On the talent front, most firms are taking proactive measures. Sheth said that at Gupshup, 35 per cent of coding workflows are already AI-augmented. "We offer hands-on training, online modules, and ensure cross-functional fluency."

Ramamurthy noted that Contentstack has embedded real-world pilot projects, internal labs, and AI adoption hackathons. "We're preparing teams not just to understand Agentic AI but to shape it."

Rathi added, "We're focused on systems thinking and human-AI cooperation. Training alone isn't enough—we need a mindset of experimentation and agility."

Economic impact across sectors

The potential of Agentic AI to contribute to India's GDP is not just aspirational—it's already visible in pilots across healthcare, agriculture, and education.

In healthcare, Gopalan highlighted how multilingual voice agents are reducing patient no-shows and improving rural care delivery. Sen added that AI triage agents could handle up to 60 per cent of OPD cases—critical in a country with a 1:800 doctor-patient ratio.

In agriculture, adaptive agents are already helping farmers optimise sowing schedules and predict pest outbreaks. According to Parab, this has led to 15 per cent higher yields and 20 per cent water savings in early pilots.

In education, voice-powered tutors and agent-based learning systems are helping students in Tier-2 and Tier-3 cities study at their own pace. As Rathi said, "This isn't just about innovation—it's about equitable access."

Parab also underscored the macroeconomic angle. He said, "If executed correctly, Agentic AI could add USD 300 billion to India's GDP by 2030."

India's entry into the Agentic AI era is not merely speculative, it is already underway. However, true readiness will depend on holistic progress. As Namburu puts it, "The promise of AI lies in making it truly work for people. That's the essence of transformation."