As consulting firms race to fold artificial intelligence (AI) into their client offerings, Deloitte India is betting on what it calls the “agentic” future of AI. At its Coalesce 2025 event on Wednesday, the firm announced a multi-year collaboration with Amazon Web Services (AWS) to establish India’s first Agentic AI Lab in Bengaluru and an AWS Centre of Excellence.
The partnership, part of a global strategic collaboration agreement, aims to expand Deloitte India’s cloud business by ₹2,500 crore by 2030 and create industry-specific AI use cases across financial services, manufacturing, consumer goods and the public sector.
Deepak Kagliwal, executive director at Deloitte India, speaks with Forbes India about how the firm defines “agentic” AI, what the new lab will focus on, and how it plans to address the growing scrutiny around AI governance. Edited excerpts:
Kagliwal describes “agentic” systems as the next phase in enterprise automation. Unlike predictive or prescriptive AI models, these can execute defined tasks within business processes instead of waiting for human input. “Earlier, an algorithm gave you an outcome, but you had to decide what to do with it. Now the agent can use that outcome to take a step forward,” he explains. “If an invoice is clear and has all the required details, the agent can directly approve and process it instead of waiting for an individual.” This, he adds, is driving a sharp rise in enterprise interest. “That is the difference. That is why demand for agents has gone through the roof.”
The new lab will serve as a development hub for both cross-functional and sectoral applications of AI. “We have a horizontal track for functions like human resources and procurement, and industry-specific tracks across financial services, manufacturing, consumer and public sector,” Kagliwal says.
Deloitte plans to draw on learnings from its global network of AI labs in the United States, Canada and the Asia-Pacific region. The India lab will begin with a 10-15-member team comprising technical and industry experts, and expand as project demand grows.
According to Kagliwal, the firm has already used agentic AI in client projects. “For a large steel company, we used Amazon Bedrock agents to convert legacy code into cloud-native applications. The project lifecycle, which used to take six months, can now be completed in about a month,” he says.
For a banking client, Deloitte has deployed AI agents to automate the entire development lifecycle—from requirement documentation to coding and testing. “Every phase now has agents doing their part of the work and reducing effort,” he adds.
Also Read: How can companies secure their future with rise in agentic AI adoption
The collaboration builds on a decade-long global alliance between Deloitte and AWS. Commenting on the India initiative, Praveen Sridhar, Head of Partner Business, AWS India and South Asia, said, “With Deloitte, we are making it easier for enterprises and the public sector to adopt cloud-driven solutions that address real business challenges, whether it is improving customer service, streamlining operations or driving sustainable growth. The Deloitte AWS Centre of Excellence and the new AWS Agentic AI Lab will give organisations across India the tools and expertise they need to reimagine what is possible with AI on the cloud.”
While Deloitte expects AI to improve speed and efficiency, Kagliwal says adoption is now being shaped by client oversight. “When we deliver, clients want us to use AI to make sure we are delivering faster and at a lower cost,” he says. “Traditionally, if I code manually versus using an AI assistant, my time to market and cost should come down by at least 30 percent. That is the expectation.”
However, he adds, AI tools cannot be deployed until clients approve them. “Every client is at a different maturity level. A bank, for example, will not allow any AI assistant until its information-security team has cleared that specific system for coding.”
When asked about the incident in Australia, where Deloitte’s local affiliate refunded part of a 440,000-dollar consultancy fee after admitting that a government-commissioned report contained AI-generated errors and fabricated citations, Kagliwal did not comment on that case directly, but reaffirmed that Deloitte India operates under stringent security and governance protocols.
“Every firm has its own safeguards,” he says. “Deloitte takes security with utmost care and responsibility. That is the core of the business we do, and there are robust mechanisms in place to ensure compliance.”
The Australian government has since indicated that future consultancy contracts may include stricter AI-usage clauses.
Kagliwal adds that in India, projects undergo multiple layers of scrutiny and clients also approve the models used. “We follow Deloitte’s Trusted AI framework that takes care of governance, data privacy and security. Clients are also directly involved in approving the models we use. Every project goes through their own information-security process before deployment,” he says.
Kagliwal says the collaboration will also drive talent development in India, creating opportunities across cloud, data, AI, SAP (systems, applications and products in data processing) and Internet of Things (IoT) skills. Deloitte plans to combine external hiring with internal upskilling through continuous training and certification in partnership with AWS.
He believes India is well placed to contribute to global innovation. “Through the IndiaAI Mission, the country is building data infrastructure and even basic large language models,” he says. “We are behind some others, but we are catching up fast. A good example is Bhashini, which supports multiple Indian languages and is empowering public sector use.”
For FY25, the Deloitte reported global revenue of $70.5 billion, a 4.9 percent increase over the previous year. The firm’s global workforce also rose to more than 470,000 people, with continued hiring across cloud, data and emerging technology roles. Deloitte has announced plans to invest over $3 billion in generative AI through FY30 to enhance its delivery capabilities, improve productivity and develop new product offerings.
While Deloitte does not disclose country-wise revenue, at the end of September, Romal Shetty, CEO of Deloitte South Asia, told the Press Trust of India that the firm is targeting a fourfold increase in revenue from its India business to $5 billion (around ₹40,000 crore) by 2030, aiming to become the “undisputed leader” in professional services. He added that Deloitte India’s revenue crossed the ₹10,000 crore mark during the June 2023-May 2024 period, reflecting a 30 percent year-on-year growth and contributed about 10 percent to Deloitte’s global growth.
The firm also hired more than 25,000 employees last year, taking its India headcount to about 1.2 lakh. Attrition fell to 13 percent, below the Big Four average.
First Published: Nov 12, 2025, 14:01
Subscribe Now