Huge opportunities for tech companies in AI age, but transition will be painful: HCL Tech MD & CEO
Stresses on the need for massive reskilling of the workforce in specialised skills.
- Feb 24, 2026,
- Updated Feb 24, 2026 3:31 PM IST
Over the past few days, there has been a lot of debate and discussion around how artificial intelligence (AI) deployment and its increasing use cases could perhaps destroy the software services industry that employs millions of people and generates billions of dollar of revenue.
C. Vijayakumar, the MD and CEO of HCLTech, stressed this was not the time to write the sector’s obituary. In fact, he sees significant new opportunities ahead for the industry driven by AI adoption. However, he feels it will be a “painful transition” and called for a massive reskilling of the work-force.
“Whether it is cloud, digital, Y2K, all of that needed more people. This time, a lot of what our people do can be done a lot more efficiently with significant speed. So, I would say this transition is different from the other transitions. It's going to be painful, because it really involves people,” said Vijayakumar.
Today AI is not the problem. But it is the ability to scale the adoption of AI in enterprises to deliver measurable outcomes that is really the challenge, he feels. He stated that the they were “disproportionately investing” in creating IPs that would help the adoption of AI.
“The biggest opportunity lies in making this intelligence layer relevant to help enterprises scale AI adoption,” said Vijayakumar speaking at the Nasscom Technology and Leadership Forum in Mum-bai.
He, for instance, pointed out that there is a trillion-dollar capex which is going into building AI data centres across the world and that will mean at least $100 billion of services spent in designing, implementing, managing data centres on an ongoing basis, which is a big opportunity.
“Physical AI is a huge opportunity. It's believed to be $1 trillion. And I think that will at least mean $100 billion of services spent in building solutions, building proof points and deploying them across multiple industries,” pointed Vijayakumar.
He believes the recent market reaction that led to a huge sell off in shares of software companies was overblown. He pointed the real world had lot of complexities and companies couldn't just plug and play AI replacing legacy systems.
Also, according to him, there is a big lag between how fast the technology is evolving and how it is getting deployed in enterprises, and that's where the challenges are going to be.
“If all the fantastic AI technologies and frontier models need to land and deliver value to an enter-prise, it's the service companies which can really deliver,” said Vijayakumar.
He stressed that while the core software engineering concepts will remain relevant, demand for specialised skills in areas like cloud, AI, data and security was going to increase significantly.
“I think we need to focus tremendously on reskilling and reskilling is all about how can we get every one of our engineers to really become a super user of AI,” noted Vijayakumar.
He pointed that while tools could be used to create codes faster, none of it could be deployed with a human review.
Over the past few days, there has been a lot of debate and discussion around how artificial intelligence (AI) deployment and its increasing use cases could perhaps destroy the software services industry that employs millions of people and generates billions of dollar of revenue.
C. Vijayakumar, the MD and CEO of HCLTech, stressed this was not the time to write the sector’s obituary. In fact, he sees significant new opportunities ahead for the industry driven by AI adoption. However, he feels it will be a “painful transition” and called for a massive reskilling of the work-force.
“Whether it is cloud, digital, Y2K, all of that needed more people. This time, a lot of what our people do can be done a lot more efficiently with significant speed. So, I would say this transition is different from the other transitions. It's going to be painful, because it really involves people,” said Vijayakumar.
Today AI is not the problem. But it is the ability to scale the adoption of AI in enterprises to deliver measurable outcomes that is really the challenge, he feels. He stated that the they were “disproportionately investing” in creating IPs that would help the adoption of AI.
“The biggest opportunity lies in making this intelligence layer relevant to help enterprises scale AI adoption,” said Vijayakumar speaking at the Nasscom Technology and Leadership Forum in Mum-bai.
He, for instance, pointed out that there is a trillion-dollar capex which is going into building AI data centres across the world and that will mean at least $100 billion of services spent in designing, implementing, managing data centres on an ongoing basis, which is a big opportunity.
“Physical AI is a huge opportunity. It's believed to be $1 trillion. And I think that will at least mean $100 billion of services spent in building solutions, building proof points and deploying them across multiple industries,” pointed Vijayakumar.
He believes the recent market reaction that led to a huge sell off in shares of software companies was overblown. He pointed the real world had lot of complexities and companies couldn't just plug and play AI replacing legacy systems.
Also, according to him, there is a big lag between how fast the technology is evolving and how it is getting deployed in enterprises, and that's where the challenges are going to be.
“If all the fantastic AI technologies and frontier models need to land and deliver value to an enter-prise, it's the service companies which can really deliver,” said Vijayakumar.
He stressed that while the core software engineering concepts will remain relevant, demand for specialised skills in areas like cloud, AI, data and security was going to increase significantly.
“I think we need to focus tremendously on reskilling and reskilling is all about how can we get every one of our engineers to really become a super user of AI,” noted Vijayakumar.
He pointed that while tools could be used to create codes faster, none of it could be deployed with a human review.