IBM’s Vishal Chahal on how enterprises can maximise value from AI agents

As artificial intelligence (AI) continues to evolve from basic automation to autonomous decision-making, agentic AI – AI that can perform complex tasks with little or no human intervention – is rapidly gaining traction. Software giant IBM, which is betting big on this AI system, recently launched an Agentic AI Innovation Centre in Bengaluru to facilitate enterprise co-creation and experimentation with autonomous AI agents. In a recent interview with TechCircle, Vishal Chahal, Vice President, IBM India Software Labs, explains agentic AI's transformative potential and IBM’s role in this process, specifically the new Agentic AI Innovation Centre in India and its upcoming plans. Edited excerpts.

How are AI agents evolving, and what impact is it having on the AI development lifecycle?

AI agents are rapidly evolving due to advances in AI tooling, integration frameworks, and modular platforms. This evolution of what we call agentic AI centres on scalable agent orchestration, integration with enterprise systems, and robust oversight mechanisms. These advancements are moving enterprises from prototyping to full-scale AI deployment. Composable platforms enable developers and business users to collaboratively build workflows using low-code and no-code interfaces. AI contributes to up to 30% of new code development, accelerating innovation and allowing human teams to focus on strategic functions.

How do you see Agentic AI unlocking new opportunities for the enterprise? When will CXOs see its real benefits?

Agentic AI is transforming enterprise operations by automating processes, improving decision-making, and delivering ROI. These intelligent digital collaborators streamline tasks across HR, IT support, customer service, and supply chain management. IBM's experience as "Client Zero" demonstrates this potential, with $3.5 billion in cost savings across 70-plus AI-integrated workflows. Agentic AI also empowers non-technical users to drive automation. CXOs will see substantial benefits in the next two to three years as enterprise-scale deployments stabilise, boosting productivity, resilience, and agility.

What are the specific concerns for companies when it comes to agentic AI deployment?

Agentic AI deployment raises governance, scalability, and trust concerns. Enterprises must implement stringent security and access controls from the outset. Managing potentially thousands of parallel agents through "agent ops" is crucial. Human oversight is essential, particularly for critical decisions requiring interpretability in sectors like finance and healthcare. Integrating new AI systems with legacy infrastructure across fragmented environments is another key challenge, requiring data readiness and architectural solutions before scaling.

What sets IBM’s strategy apart in building responsible AI agents?

IBM’s responsible Agentic AI strategy prioritises trust, security, and governance. IBM views AI agents as collaborators that augment human decision-making, maintaining a human-in-the-loop model for transparency. Ethics are integrated into core tools like WatsonX Orchestrate, aligning with enterprise security frameworks and global regulations. The Bengaluru software lab refines these capabilities, focusing on agent management and governance. This approach ensures effective, accountable, and explainable AI deployments aligned with organisational standards.

Where does India fit in IBM’s global software and AI innovation strategy?

India is central to IBM’s global AI and hybrid cloud strategy, leveraging its talent and R&D hubs. The IBM India Software Lab (ISL) contributes to design, engineering, development, service, and support for all major product lines. It's the only IBM lab working across multiple platforms, delivering solutions in automation, data & AI, security, and sustainability. Beyond product development, IBM’s India labs collaborate with the local tech ecosystem, including academia, startups, governments, clients, and partners, for innovation, exemplified by the Agentic AI Innovation Centre in Bengaluru.

What is the significance and plans for the new Bengaluru centre?

The Agentic AI Innovation Centre in Bengaluru enables companies to transform AI concepts into practical solutions. It gives clients, partners, and developers a space to experiment, learn, and build with IBM’s AI tools. The focus is on real business use cases like automating tasks, managing AI systems at scale, and improving productivity. Teams will have access to tools to train small language models, set up AI workflows using Watsonx Orchestrate, and audit AI systems more efficiently. Over time, the centre will also work on strengthening AI governance, integration, improving how agents communicate, and helping organisations manage complex AI deployments.

What are IBM's plans for the future of enterprise AI, both globally and in India?

Globally, IBM is leading in operationalising AI at scale through Agentic AI systems that enhance productivity and human-AI collaboration. Platforms like WatsonX orchestrate and empower organisations to automate workflows while embedding security, compliance, and explainability. These systems integrate with existing IT ecosystems, promoting trusted AI adoption. IBM addresses adoption barriers, such as unstructured data, by simplifying the data-for-AI pipeline with WatsonX data. In India, this vision is realised through R&D and ecosystem-building. IBM’s India labs develop global technologies and foster local innovation, such as the collaboration with IIT Kanpur to create AI solutions for air pollution. These efforts shape a future of trustworthy, scalable enterprise automation.