AI Isn't Just About Speed, Efficiency Isn't Enough: Anthropic India MD

India isn't a small market to solve for. It's over a billion people. Irina Ghose said scale is exactly where AI can prove its worth.

10 Sep 2026, 09:16 · NDTV

Show
Quick Read
Summary is AI-generated, newsroom-reviewed

Irina Ghose says AI efficiency is only half the job; innovation matters more
Leading firms track AI efficiency and how saved time fuels bigger goals
AI increasingly works invisibly, delivering outcomes without user friction

How should companies measure innovation instead of just efficiency?

Companies love to talk about AI efficiency. Faster reports. Quicker code. Fewer hours on boring tasks. But Irina Ghose, Managing Director of India at Anthropic, thinks that's only half the job.

Speaking to NDTV Profit on the sidelines of the Global Fintech Fest in Mumbai, Ghose said businesses need to look past the speed AI gives them. The real question, she said, is what they do with the time it frees up.

"Efficiency is bringing in headroom for capacity for innovation," Ghose said.

Also Read | Layoffs Are Rising In Tech, But These Skills Are Helping People Unlock Better Pay

Two Things To Track

Ghose said the companies leading on AI right now are watching two numbers, not one. First, how much efficiency AI is creating in day-to-day work. Second, how that saved time and effort is being redirected toward bigger goals.

She said firms should pick their "big rocks or the large moonshots" and measure progress against them, the same way they measure efficiency. Not every company is at this stage. Some have gone all in on AI. Others are still figuring out timing and approach. For the ones already using it, efficiency is the easy metric to point to.

But that's where most stop, according to Ghose. "Now, are you really measuring the innovation?" she asked.

This point echoes what industry watchers in India have been saying too. Shrish Anand Lal, Executive Director & Chief Business Officer at New Street Technologies, told NDTV, "Efficiency gains plateau fast, you can only make an existing process so much faster. The real shift is enterprises moving from AI that assists a workflow to AI that configures the system the workflow runs on, with humans approving and deterministic systems executing. That is where AI stops being a cost line and starts being how the business actually operates."

AI You Don't Even Notice

Ghose also talked about a shift already underway inside companies: AI is quietly disappearing into the background. "As long as you think of AI as something which is doing outside of you, there's still a little level of a gap and a friction," she said.

She gave a simple example. Someone reviewing a month's financial reconciliation doesn't need to see the AI at work. They just see the finished reconciliation, with the logs and checks already done behind the scenes.

A data analyst might work the same way -- caring about the insight, not the steps that built the slide deck around it. "Only the outcome matters," Ghose said.

The Model Isn't The Whole Story

Ghose was clear that a good AI model alone doesn't guarantee results. What a company builds around that model matters just as much, maybe more.

"The model is a game changer, but the capability of the builder who's building things with the model is what defines and actually defines the entire change in the market," she said.

She pointed to real examples in India. Cognizant, she said, has rolled out AI to 350,000 employees. Axis Bank is testing it in engineering work. IndusInd Bank is exploring how it can change knowledge work altogether.

Ghose said AI is especially useful where companies are stuck with old, tangled systems that don't talk to each other well.

In a similar vein, Ankesh Kumar, Director Marketing, Ingram Micro India, told NDTV, "AI's impact will go far beyond efficiency. For many organisations, productivity may be the first use case, but the larger opportunity is how AI reshapes business models, customer experiences, decision-making and the way entire ecosystems operate."

India's Scale Is The Real Test

India isn't a small market to solve for. It's over a billion people, hundreds of languages, and huge gaps between cities and villages. Ghose said that scale is exactly where AI can prove its worth.

Anthropic, she said, is working to improve AI performance across more than ten of India's most widely spoken languages. The goal is to make the technology usable well beyond English and beyond big cities.

She flagged agriculture, education, healthcare and skilling as sectors where this could matter most. "The minute you kind of try to solve population-scale problems by making it available to the last mile, that's where the real difference lies," Ghose said.

She also mentioned ongoing work on where AI data is stored and processed. Bringing that infrastructure closer to Indian users, she said, could open doors for the country's broader tech ecosystem.

It Starts At The Top

For companies still on the fence about AI, Ghose said leadership sets the tone. Her advice: leaders should use the tools themselves, not just approve budgets for them. And they should start with the outcome they want, not with a list of AI features.

"The biggest problem, which they are not asking what can AI do, but says what good outcome would look like," Ghose said. Once a company knows the outcome it's chasing, applying AI becomes a much clearer, more focused exercise, she said. "I think that's the way the entire culture will change," Ghose said.

Got a follow‑up question on this article? Go on

How may i help you today

Show full article

Track Latest News Live on NDTV.com and get news updates from India and around the world