Ramprakash Ramamoorthy, Director - AI Research, Zoho Corporation.

AI is improving enterprise productivity by roughly 20-25 per cent rather than triggering the mass replacement of white-collar workers predicted by industry hype, according to a Zoho Corporation leader.

“I have heard commentary about how it would take 18 months to replace all white-collar jobs. We are yet to see the levels of productivity both internally and with our customers. At best, we see 20 to 25% improvement in productivity, which is not enough to replace human involvement. For instance, if a customer support engineer is looking at solving 10 tickets a day, you enable them to solve 12 tickets a day. That is the level of productivity,” said Ramprakash Ramamoorthy, Director - AI Research, Zoho Corporation.

While software development has delivered some of the highest productivity gains from AI, there is still a significant amount of work ahead, with at least five years’ worth of products and capabilities to be built on the roadmap.

He noted that one possible driver behind recent global layoffs could instead be the rising cost of infrastructure and compute required to support AI deployments.

“To buy the same server we purchased four months ago for our data centre operations now costs nearly twice as much, largely because AI demand is driving up chip prices,” Ramamoorthy said. “Chips, memory, hard disks, firewalls, and routers — almost every component has become nearly twice as expensive.”

He added that privately held companies like Zoho Corporation are better positioned to absorb rising AI infrastructure costs because they can take a longer-term view of investments. However, companies cannot immediately pass on sharply higher compute and infrastructure expenses to customers through steep price hikes.

“With infrastructure and compute already among the biggest expenses after human resources, if compute costs suddenly double, you cannot simply raise customer pricing by 30-40 per cent overnight,” he said.

Ramamoorthy said AI has historically gone through cycles of “over-promising and under-delivering,” adding that while the technology has significantly evolved in recent years, it should be viewed primarily as a productivity enhancer rather than a doomsday disruptor.

Enterprises adopting AI would benefit more from setting practical and measurable goals instead of chasing unrealistic expectations such as exponentially scaling revenues overnight.

“We are a 19,000-person company with multiple operations. We are trying to bring AI into our workflows and enable our employees to start using AI. With customers, we are increasingly seeing that almost every request for proposal across companies of different sizes, from small businesses to enterprises, now includes sections on AI capabilities and privacy considerations around training these AI models. That tells us there is a lot of interest in adopting AI as a nation,” he observed.

Ramamoorthy said India is witnessing significantly higher energy around AI adoption compared to many other global markets, with emerging economies such as the Middle East, Southeast Asia, and Latin America also seeing accelerated uptake.

According to him, these regions have been able to adopt AI faster because many businesses were built directly on cloud infrastructure and therefore do not face the burden of migrating legacy systems.

Ramamoorthy also shared that the industry is increasingly prioritising AI deployment in “profit centres” over “cost centres”. Companies are investing more heavily in AI for functions such as CRM because better AI-enabled systems can help improve sales and generate additional revenue.

Meanwhile, adoption has been comparatively lower in functions such as email and internal chat, which are viewed more as cost centres. Across the industry, technologies that directly help companies make more money are receiving greater priority than those focused primarily on cost optimisation.

Published on May 29, 2026