AI boom could make capital costlier: Artificial intelligence is often discussed as a story about computing power, software and semiconductors. There is another side to the AI boom that is receiving less attention. Building the infrastructure needed to run increasingly powerful models requires vast amounts of capital. Data centres have to be built, semiconductor capacity expanded, electricity generated and transmitted, and cooling and other supporting systems put in place. Much of this investment will have to be financed.
That comes at a time when governments are already borrowing heavily. The combination could keep demand for capital high for longer and make the international financing environment more difficult for countries such as India.
READ | Specialised skills: In the age of AI, depth is becoming a career advantage
For much of the past two decades, the global cost of borrowing was closely tied to the monetary policies of the major central banks. When the US Federal Reserve raised interest rates, Treasury yields generally moved higher, the dollar strengthened and emerging markets came under pressure from capital outflows and higher external financing costs. Easing cycles tended to relieve some of that pressure.
The world is entering a different phase. Public debt has risen sharply across major economies, while governments and companies are embarking on investment programmes that require large amounts of capital. AI is adding another major claim on the available pool of savings.
Government bonds matter in this equation because their yields influence borrowing costs throughout the financial system. The IMF estimates that global public debt reached nearly 94% of GDP in 2025 and could reach 100% by 2029. When yields on major sovereign bonds rise, investors can demand higher returns from corporate and emerging-market borrowers as well.
For India, the issue is not that money will suddenly disappear from international markets. It is that India may have to compete harder for it. The country needs capital for infrastructure, manufacturing, urbanisation and the energy transition at precisely the time that advanced economies are seeking investment for AI, semiconductors, defence and energy security.
The financing burden of the AI boom
The investment required for AI does not end with the technology companies developing the models. The boom is creating demand for data centres, electricity generation, transmission networks, construction, semiconductor plants and power equipment. The companies involved in supplying this infrastructure will themselves need capital.
That distinction is important. An investment boom financed largely from retained earnings places a different demand on financial markets from one that relies heavily on borrowing. When companies raise debt to build capacity, they are competing with other borrowers for investors’ money.
There is already evidence of this happening. Research by the Bank for International Settlements shows that major technology companies have increasingly turned to debt markets to finance AI-related infrastructure. The five largest hyperscalers are expected to spend more than $1 trillion on capital expenditure in 2025 and 2026, while bond issuance by large technology companies rose sharply in 2025.
READ | Generative AI has changed the meaning of university degree
The effect on global interest rates should not be overstated. AI investment is one factor among several affecting the demand for capital, and higher spending by technology companies does not automatically translate into higher sovereign yields. But the scale of the investment cycle means that it can influence financial conditions, particularly when it coincides with heavy government borrowing.
The transmission to India would be through financial markets. An investor deciding between a US Treasury bond and an Indian government bond considers the return available from each after accounting for risk, currency movements and other factors. If yields on US government debt remain attractive, Indian borrowers may have to offer competitive returns to retain the interest of global investors.
That can happen even when the Reserve Bank of India is reducing interest rates. Domestic monetary policy has considerable influence over borrowing conditions in India, but it cannot completely insulate the country from movements in global bond markets.
There is another development that could keep pressure on the cost of capital. Countries are increasingly willing to accept higher production costs in exchange for greater control over strategically important industries.
For several decades, companies could make a strong case for locating production wherever costs were lowest and supply chains were most efficient. National security considerations have changed that calculation. Governments are encouraging domestic production of semiconductors, critical minerals, energy equipment and defence technologies even when doing so is more expensive.
This shift has economic consequences. Building duplicate capacity in several countries requires more investment than concentrating production in the most efficient locations. Supply-chain diversification can improve resilience, but it can also increase the amount of capital required to produce the same goods and services. If the resulting investment is accompanied by higher fiscal deficits or persistent inflationary pressure, borrowing costs may remain higher than they otherwise would have been.
For emerging economies, the timing is awkward. India wants to attract investment into electronics, semiconductors, clean energy and digital infrastructure. It is competing for that capital with countries that are themselves offering large incentives to attract strategic industries.
READ | Generative AI is forcing universities to rethink assessment
India has a stronger starting point
India nevertheless enters this period in a better position than many emerging economies. Public investment in infrastructure has remained strong, and there are signs that private investment is beginning to recover after a prolonged period of subdued capital expenditure.
The Economic Survey 2025-26 points to an improvement in private investment intentions, supported by higher capacity utilisation and a rise in new project announcements. Continued public capital expenditure has also helped create conditions in which private companies are more willing to invest.
That matters because an economy that can finance a larger share of its investment from domestic savings is less exposed to swings in international capital flows. Foreign investment will remain important, particularly for sectors requiring technology and large upfront investment, but it should complement domestic capital formation rather than substitute for it.
Foreign direct investment has an additional advantage over portfolio capital. A company building a factory or establishing a long-term supply chain is making a commitment that is harder to reverse in response to a change in global bond yields. Portfolio investors, by contrast, can shift allocations much more quickly when the relative returns on different markets change.
Recent data nevertheless show that foreign capital flows can move in both directions. Net FDI inflows were $6.1 billion in the first quarter of FY27, according to the Ministry of Finance, compared with $5.2 billion a year earlier. Net portfolio investment recorded an outflow of $9.6 billion during the same quarter.
The more important test for India is whether investment generates sufficiently high returns to compensate investors for the risks of operating in the country. That requires improvements that rarely attract the same attention as large investment announcements: dependable power, efficient logistics, predictable regulation, faster resolution of commercial disputes and deeper sources of long-term domestic finance.
India cannot determine the level of US Treasury yields or the amount of capital that global investors allocate to emerging markets. Nor can it assume that lower domestic interest rates will always produce an equivalent reduction in the cost of capital faced by Indian companies.
What India can influence is the productivity of the capital it attracts.
That consideration becomes more important as the global economy enters a period of heavy investment. AI infrastructure is likely to absorb enormous sums, while governments are spending more on energy security, defence and strategic manufacturing. The energy transition requires another large wave of investment. None of these demands is likely to disappear soon.
The AI boom could therefore have an unintended financial consequence. The technology may eventually raise productivity enough to justify the investment being made today, but getting to that point requires a huge expansion of physical capacity. Financing that expansion will compete with other uses of global savings.
For India, the implication is less about the AI industry itself than about the financial environment it could help create. The period when emerging economies could count on abundant and inexpensive international capital should not be treated as a permanent feature of the global economy. India will need to make each unit of capital more productive, while creating conditions that persuade investors to stay for the long term.