The money flowing into artificial intelligence (AI) far exceeds the revenue it currently generates, Nikhil Kamath has said. The Zerodha co-founder warned that 2027 could be a difficult year for startups and said “it won't be a bad idea” for entrepreneurs to raise funds now as access to capital could become difficult in the months ahead.
Speaking at the NDTV World Summit 2026, Kamath said the AI industry was attracting enormous investments but had yet to generate enough revenue to justify the spending.
“The amount of money going into AI and the money coming out of AI, there is a big gap in between that,” he said.
Kamath estimated that around $1 trillion was being invested in AI annually, while revenue generated through enterprise adoption was between $100 billion and $200 billion. This leaves a gap of $700 billion to $800 billion a year that investors or other sources of funding must cover, he said.
He noted that foreign capital, including investments from the Gulf, helped fund American AI companies in the previous funding cycle. Finding enough fresh capital to sustain investments at the current scale every year could prove difficult.
ALSO READ | Barclays to cut 5,000 jobs globally this year as it ramps up AI adoption
Why 2027 could be difficult for startups
Kamath said investors were counting on AI to boost productivity and generate trillions of dollars in economic benefits to justify the huge spending. “I don't know if I've seen that happen on the ground yet," he said.
He also warned that rising US Treasury yields could make borrowing more expensive. "So for next year, with American Treasury rates going up, cost of money going to 5-6-7%, I don't know where it'll go, it's at 5% today.”
“Somebody has to bridge this gap up until that five trillion dollar revenue story plays out, even if it does or it doesn't, nobody knows,” he said.
Kamath advised founders to secure funding while investor appetite remained relatively strong. “So if somebody is able to raise a lot of capital today and keep it, probably not a bad idea,” he said.
Kamath is the co-founder of Zerodha, one of India's largest stockbroking firms. He has also invested in several businesses, including Licious, Nourish You, Pee Safe, Subko and electric scooter maker Ather.
He is involved in philanthropy and environmental initiatives and has signed the Giving Pledge, a commitment to donate at least half of one's wealth to charitable causes during one's lifetime or through a will.
Kamath backs inheritance tax
At the summit, Kamath also backed the idea of an inheritance tax, calling it “probably not the worst idea in the world”.
An inheritance tax is levied on beneficiaries who receive money, property or other assets after someone's death. Kamath questioned whether wealth passed between generations should be treated differently from income earned during a person's lifetime.
ALSO READ | Nikhil Kamath marks Zerodha’s 16th birthday with nostalgic note: ‘Hope we’re all still in same room in 2042’
“I don't like the idea of taxing somebody who is earning money in his lifetime and incentivising that but taxing people between generations when wealth is being passed through; I don't know in what manner. It has to be done in a manner where it's not an unrealized gain,” he said.
He said that such a tax should not force business owners to sell equity in their companies to meet their tax obligations. He also recalled facing online criticism when he had previously shared his views on inheritance tax.
‘Wage growth must keep up with asset prices’
Kamath also discussed the Gini coefficient, a measure of income or wealth inequality. He said inequality was worsening in many parts of the world and argued that the wealth gap would be difficult to reduce unless wages grew alongside asset prices.
“As long as wage growth in society is not going to keep up with asset price inflation, logically it's impossible for wealth inequality to come down,” he said, adding that governments might need to intervene to address the problem.
He also asked wealthy individuals to take steps to reduce inequality before it becomes more difficult to tackle.
“To the rich people in this room, it's not even altruism, but it could be self-serving to solve this problem when you can, before you reach a point where you have to,” he said.
What India can learn from China on AI
Kamath said India had the potential to build major technology companies but lacked the risk capital available to businesses in the United States and China, particularly those developing advanced AI models.
He said companies working on frontier AI models had access to greater funding to experiment with new technologies, making it difficult for Indian firms to compete at the same level.
“I don't see how an Indian company, or for that matter any company outside of those two or three countries where there is that much risk capital, can have a fair shot at competing in that layer,” he said.
However, he pointed to India's cost advantages in setting up businesses and building data centres. Kamath estimated that setting up a megawatt of capacity in India could cost 30% to 40% less than in the West. He also said electricity was significantly cheaper in India than in several other countries.
Kamath suggested that India could learn from the competition between Chinese provinces, where local leaders compete to attract investment and promote development.
“One thing that maybe we can change and learn from China is to allow for different provinces or states to compete with one another like they were a capitalistic organisation itself,” he said.
He added that chief ministers could operate more like corporate CEOs, competing to improve their states' performance and attract businesses.