Agentic AI Adoption and Spending Accelerates in India India is rapidly embracing agentic AI but challenges around infrastructure, data governance, and cybersecurity still pose hurdles to responsible, scalable deployment

India is witnessing a sharp rise in artificial intelligence (AI) adoption, particularly in agentic AI, with spending expected to soar from USD 2.1 billion in 2023 to USD 10.4 billion by 2028, according to IDC and UiPath latest report, Agentic Automation: Unlocking Seamless Orchestration for the Modern Enterprise. This represents a robust compound annual growth rate (CAGR) of 38 per cent.

Currently, 40 per cent of organisations have already adopted agentic AI, with C-suite leaders expecting a 3–4x return on investment within the first year. Another 56 per cent plan to adopt agentic AI within the next two years, and 46 per cent expect to roll it out within the next six months.

Despite the enthusiasm, most companies remain in the early stages of deployment. The IDC report notes that 64 per cent of organisations are still conducting proof-of-concept (PoCs) without defined budgets, while only 19 per cent have clear spending plans in place.

Among the sectors leading the charge are manufacturing, retail and wholesale, and healthcare and life sciences.

However, realising the full potential of agentic AI is not without its challenges. Enterprises face several hurdles, including biased datasets, complex data engineering, and outdated IT infrastructure.

"Ensuring responsible and scalable deployment of agentic AI starts with building the right foundation," DebDeep Sengupta, Area Vice President, South Asia, UiPath emphasised. "Organisations need to invest in high-quality, diverse datasets and implement governance frameworks that actively monitor for bias and promote ethical AI use. In a market as diverse as India, data integrity cannot be an afterthought."

On the infrastructure front, Sengupta noted the need to modernise legacy systems and adopt flexible, cloud-native platforms that can support real-time AI workloads. "The goal is to enable secure, seamless interaction between AI agents, internal systems, and third-party models with full transparency."

Security and talent remain key concerns

Despite the promising benefits, businesses are wary of cybersecurity threats and ethical risks. According to the report, 53 per cent of organisations are concerned about data privacy breaches, 47 per cent about lack of transparency in AI decision-making, and 46 per cent about potential misuse by malicious actors.

Implementation challenges are also notable. The lack of skilled IT talent (53 per cent), data security issues (47 per cent), and regulatory uncertainty (47 per cent) continue to be major bottlenecks for Indian enterprises.

Sengupta advocates for a dedicated trust layer to ensure transparency and security. This includes robust access controls, real-time analytics, and detailed reporting on AI utilisation across business processes. He also stressed the importance of internal capability-building. "Enterprises must upskill their workforce not just in AI/ML, but in automation, data engineering, and ethical AI practices as well."

"More importantly, AI literacy must extend beyond engineers to legal, compliance, and operations. Without this foundation, agentic systems risk becoming ungovernable at scale," added Tarun Dua, Founder and MD of E2E Cloud

Nevertheless, around 80 per cent of organisations report significant productivity gains, while 73 per cent value the flexibility to use multiple foundation models within a single task. A similar proportion believes that agentic AI enhances decision-making across business functions.