Archive · 5 Oct 2026
AI’s Next Phase Shifts Enterprise Focus From Experiments to Economics
Many companies have been trying out AI in small projects.
The article says the next step is to see whether AI improves business results.
Banks and insurers can use it to help with tasks like fraud detection and customer service.
Shops and online businesses can use it to help customers find products and manage stock.
Telecom companies can use AI to answer customer questions and improve their networks.
Factories can use it to reduce breakdowns and waste.
These changes may help companies earn more or spend less.
The article says investors should look for measured results, not just AI experiments.
The article argues that enterprise AI is moving from pilots toward measurable financial results.
BFSI, digital-native retail and consumer businesses, telecom, and manufacturing are identified as sectors positioned to gain.
In BFSI, AI is being applied to areas including underwriting, collections, compliance, and customer service.
Retail and consumer businesses can use AI to affect sales, pricing, fulfilment, forecasting, and inventory efficiency.
Telecom and manufacturing can use AI to improve productivity and operating performance, including network efficiency, uptime, yields, and working capital.
- Who
- Large enterprises, particularly in BFSI, digital-native retail and consumer businesses, telecom, and manufacturing.
- What
- The article argues that enterprise AI is shifting from experimentation toward measurable economic results.
- When
- The article describes this as the next phase of enterprise AI.
- Where
- The article discusses sectors in India and more broadly, including examples from Indian businesses.
- Why
- AI’s value, the authors argue, depends on whether it improves revenue, costs, working capital, or return on capital.
- Sectors identified as ahead
- BFSI; digital-native retail and consumer businesses; telecom; manufacturing
- BFSI use cases
- Underwriting, collections, compliance, customer servicing, cross-selling, wealth management, and software development
- Retail and consumer use cases
- Discovery, search, recommendations, pricing, promotions, inventory, fulfilment, fraud, and customer service
- Telecom example
- A large Indian telecom operator’s GenAI assistant handles millions of customer interactions monthly.
- Potential telecom productivity gain
- The article says a 10–20% productivity improvement could be material to EBITDA.
- Manufacturing use cases
- Computer vision, predictive maintenance, supply-chain forecasting, digital twins, robotics, and physical AI
- Authors
- Karan Taurani, EVP at Elara Capital, and Piyush Gupta, VP, APAC, Middle East, India, at VULTR