India’s manufacturing giants are embracing agentic AI to enhance efficiencies

The manufacturing sector, is traditionally defined by the precision of heavy machinery and the grit of physical labour. With digital getting into every sphere of life, the industry is moving beyond “automated” tasks towards “autonomous” decision making. At a recent industry session hosted by Salesforce, leaders from top manufacturing and chemical firms signalled that the industry is moving beyond basic robotics into the era of the Agentic Enterprise. This is a model where AI is no longer just a digital tool but a strategic teammate capable of managing complex logic.

For India’s manufacturing landscape, this means a shift from reactive troubleshooting to a seamless, ‘human-in-the-loop’ ecosystem where AI manages everything from demand forecasting to alternative supply paths.

“What is happening with agentic AI is it is giving you the option to take decisions of its own and that is extremely important. It tells me when to order, from where to order, and what are the alternative supply paths so the entire chain process becomes seamless,” he said.

This evolution also necessitates a change on the factory floor. Kumar noted that reskilling is the only “carrot” required, adding that AI usage is becoming a baseline professional requirement. “If in a whole CV you don’t have the word AI usage to your credit, I’m sorry, you are in for a bad experience,” he added.

“Forty percent of the data leakage happens, and the main source of data leak is AI only. There are a lot of risks we are inviting when we are bringing AI into the systems, especially when you integrate IT and OT (Operational Technology),” he said.

To mitigate these risks, firms are adopting a hybrid security posture. Dr Sahoo’s firm has taken the unconventional step of “going back to storage” for sensitive data, limiting laptop usage in certain plants while simultaneously deploying “robotic security” where AI and physical robots work together to monitor perimeters.

“Getting AI into the flow of work is very critical. It cannot be a tool where you open another app and then start using it. At some point, the handshake between applications is not going to be APIs, but it is actually going to be the agent-to-agent transaction,” he said. By shifting to an agent-to-agent model, manufacturers can ensure that technology supports the worker without disrupting the operational flow.

“Those companies who will work on their employees and make them enabled now will be able to deliver better customer experiences, show higher revenue growth and deliver efficient operations. We are moving from simple information retrieval to multi-agent orchestration,” she said.

The practical impact is already visible in the sales cycle. Jay Kotecha, Senior Solutions Engineer at Salesforce, spoke about how AI now handles the manual intensive process of reading Requests for Quotation (RFQs) and tailoring product portfolios.

“What if you had a set of agents who are working with your sales rep and automating the entire workflow? We have agents researching accounts, reading RFQs on the fly to help generate quotes, and automating all follow-ups to give a revenue uplift of almost 2%,” he said.

Note to Readers: This article was produced as part of Salesforce's Strategic Leadership Dialogue. Mint is a media partner.