In an exclusive interview with Rahul Koul, Editor, India Pharma Post, Duraisamy Rajan Palani discusses the evolution of Agentic AI in the pharmaceutical industry, the key challenges surrounding its adoption, and how Archimedis Digital is contributing to this transformation.

Agentic AI is now being described as the next leap beyond copilots and automation. What makes it especially relevant for pharmaceutical and life sciences workflows?

Agentic AI is particularly well-suited to life sciences because the industry is inherently process-heavy, documentation-driven, and tightly regulated. Every stage across drug development, manufacturing, and quality demands structured documentation, validation, and traceability.

While there are several applications, two areas explain the extreme use cases in our experience. First, scientific workflows such as drug discovery, where AI can now generate molecular structures, simulate toxicity, and significantly compress timelines that traditionally took years. Second, operational workflows, especially GxP validation, documentation, and compliance, which remain mandatory across the product lifecycle.

A critical reality is that nearly 60–70% of effort in regulated pharma functions like quality and regulatory affairs is still spent on documentation and compliance-related activities. This makes them ideal candidates for agentic AI. What sets agentic systems apart is their ability to move beyond assistance toward end-to-end workflow execution. When layered on structured systems like ERP / QMS platforms that capture batch records and audit trails, they enable not just automation but intelligent orchestration and decision-making.

This is what makes agentic AI powerful in life sciences—it transforms static, manual compliance processes into dynamic, skill-infused, learning systems that continuously improve quality and efficiency.

Many pharma firms are stuck in pilot mode. What separates successful enterprise-scale AI transformation from isolated proof-of-concepts?

The primary challenge is that many organisations attempt to implement AI without first establishing a strong digital foundation and data governance. Successful transformations follow a clear progression: structured knowledge graphs, real-time data integration, and only then the application of AI.

When organisations skip these foundational steps, they struggle to scale because AI is only as effective as the quality and structure of the data it relies on.

Equally important is the approach to implementation. A disciplined “seek, shape, scale” methodology ensures that efforts are grounded in real business problems, designed for scalability from the outset, and ultimately embedded into operations. Enterprise success in pharma is not about running multiple pilots—it is about integrating AI into business workflows that deliver sustained value.

From Archimedis Digital’s client engagements, where are you already seeing measurable ROI from agentic AI in India’s pharma sector?

Measurable ROI is most evident in document-intensive and compliance-heavy functions, which are central to life sciences operations. Regulatory and quality document generation and review, for instance, benefit significantly from AI by improving first-time acceptance rates through better alignment with regulatory expectations.

In clinical workflows, automating data structuring and submission readiness reduces manual effort while creating scalable systems that go beyond one-time fixes. Similarly, in quality and manufacturing, integrating AI with ERP systems reduces cycle times in batch review, enhances traceability, and minimises errors.

There is also growing impact from capabilities like synthetic data generation, which allows companies to simulate clinical studies in advance. This helps identify protocol gaps before trials begin, saving both time and cost.

Importantly, ROI in this context extends beyond financial gains, it is reflected in faster approvals, reduced rework, stronger compliance, and more informed decision-making.

What role do platforms like Archimedis’ eCapsule and compliance-focused SDLC agents play in scaling AI beyond experimentation?

Platforms like eCapsule are foundational because they address the core challenge of data fragmentation and compliance complexity in life sciences. By integrating manufacturing, quality, and supply chain into a unified system with capabilities such as electronic batch records, audit trails, and real-time tracking they create a structured, high-integrity context layer. This is essential for AI to function reliably.

On top of this foundation, AI platforms such as Lizy AI powered by domain-specific small language models enable capabilities like document authoring, GxP reviews, risk analysis, and synthetic data creation. These are increasingly evolving into multi-agent systems, where multiple AI agents collaborate to execute complex tasks more efficiently.

Together, this Archimedis stack ERP, AI platform, and domain-trained models allows organisations to move beyond isolated pilots toward enterprise-grade, compliant AI systems that can operate at scale.

What mindset shifts are needed among leadership teams to make AI transformation sustainable rather than experimental?

The most important shift is recognising that AI is not the starting point data and systems are. Leaders must prioritise building strong digital foundations data governance before expecting meaningful AI outcomes.

There also needs to be a move away from one-off problem solving toward designing scalable systems that can be embedded into core workflows. In practice, this means focusing on long-term transformation rather than short-term automation wins.

Another critical shift is toward domain-led AI. Generic models have limited effectiveness in highly specialised industries like pharma, which is why organisations are increasingly adopting domain-specific models trained on life sciences data and workflows.

Finally, leadership must embrace a collaborative ecosystem mindset. Even as companies build internal capabilities through GCCs, external partners remain essential for innovation and specialised expertise. Sustainable transformation happens when AI is treated as a core operational capability rather than an experimental initiative.

If 2025 was the year of AI pilots, will 2026 be the year Indian pharma finally operationalises enterprise-grade agentic AI at scale? What will define the winners?

2026 is likely to mark the early phase of scaled adoption  but only for organisations that have built the right foundations. The Indian pharma industry is already evolving, supported by increased investment in digital manufacturing, modernisation of facilities, and a shift toward innovation-led growth.

AI is acting as a catalyst in this transition, enabling companies to move beyond low-margin contract manufacturing toward higher-value, innovation-driven models.

The winners will be those that have invested in strong digital foundations and data governance, adopted domain-specific AI, and embedded these capabilities into core workflows such as quality, regulatory, R&D, and manufacturing. Just as importantly, they will focus on continuous, scalable solutions rather than one-time deployments.

Ultimately, the shift is from efficiency-driven operations to intelligence-driven enterprises. Organisations that make this transition will not only adopt AI successfully but also redefine their position in the global pharma value chain.