By Karan Taurani & Piyush Gupta

The first phase of enterprise AI was about experimentation. The next will be about economics. Almost every large enterprise is testing AI in some form, making the number of pilots or use cases increasingly irrelevant. The more important question is: Where is AI already moving revenue, costs, working capital, or return on capital?

On this measure, four sectors appear ahead of the curve—BFSI, digital-native retail and consumer businesses, telecom, and manufacturing. Their advantage comes from a combination of large usable data sets, digitally mature workflows, high-transaction volumes, and outcomes that can be measured financially.

BFSI is arguably the most mature AI adopter today. Banks, insurers, and fintechs have used machine learning for years across fraud detection and credit scoring. AI is now moving deeper into underwriting, collections, compliance, customer servicing, cross-selling, wealth management, and software development.

The attraction lies in the measurable ROI. Lower fraud losses, higher collection efficiency, faster underwriting, better cross-sell, and lower servicing costs all flow relatively directly into earnings. BFSI thus shows that the most valuable AI model may simply be the one most deeply embedded into a high-frequency economic workflow.

Retail and e-commerce potentially have the strongest linkage between AI and incremental revenue. AI can influence almost every transaction: Discovery, search, recommendations, pricing, promotions, inventory placement, fulfilment, fraud, and customer service. A large value-focussed Indian e-commerce platform has put AI at the centre of its operating architecture, processing billions of signals to understand customer intent and context. This matters because small improvements become powerful at scale. A modest increase in conversion or basket size multiplied across millions of transactions can materially alter revenue and profitability.

But the bigger opportunity may extend beyond P&L. AI-led demand forecasting at a large consumer company has improved forecasting accuracy and reduced finished-goods inventory. AI therefore has the potential to simultaneously improve growth and capital efficiency—higher conversion at the front end and better inventory turns at the back.

Telecom represents the cost-productivity side of the AI opportunity. Operators have millions of customers, vast networks, and large service organisations. AI can address customer service, churn, network optimisation, predictive maintenance, sales, marketing, and software development. A large Indian telecom operator’s GenAI assistant handles millions of customer interactions monthly, with a high proportion resolved end-to-end. At this scale, even a 10-20% productivity improvement can become material to EBITDA.

Manufacturing could ultimately become an even larger value pool. Here, AI moves from augmenting knowledge workers to optimising physical assets. Computer vision, predictive maintenance, supply-chain forecasting, digital twins, robotics, and physical AI can directly affect uptime, yields, energy consumption, inventory, and throughput. The financial translation is powerful: Lower inventory improves working capital; reduced downtime raises asset utilisation; better yields reduce waste; and higher throughput improves return on capital employed (ROCE).

This leads to a broader conclusion. The sectors generating the highest AI returns today are not necessarily those deploying the most advanced models—they are those embedding AI into the largest measurable economic workflows.

The question for investors should move beyond “Who is adopting AI?” That is the real transition underway—from AI experimentation to AI economics. Ultimately, the companies that convert AI spending into measurable improvements in revenue, margins, working capital, and ROCE will be the ones that create durable shareholder value.

The writers are respectively EVP, Elara Capital, and VP, APAC, Middle East, India, VULTR.

Disclaimer: The views expressed are the author’s own and do not reflect the official policy or position of Financial Express.