From FOMO to fear of losses: How investors can handle AI stock slump
A sharp correction in AI and semiconductor stocks has exposed concentration risks in overseas portfolios. Experts advise Indian investors to review overlapping holdings and rebalance allocations
)
The correction in AI and semiconductor stocks has exposed concentration risks for Indian investors (Representative image from file)
Listen to This Article
This month’s sharp correction in artificial intelligence (AI) stocks has left Rohan Mehta wondering whether to exit foreign equity funds. The 38-year-old product head at a Bengaluru software firm, who earns around Rs 50 lakh a year, invested abroad after Indian equities corrected from their September 2024 high. His dilemma mirrors that of many Indians who followed the same path.
Mehta (name changed on request) began cautiously in early 2025, with a Standard & Poor’s 500 (S&P 500) index fund. As AI stocks rallied, he increased his exposure to the AI and semiconductor exchange-traded funds (ETFs) and stocks via the Liberalised Remittance Scheme (LRS). High returns also drew him beyond the United States (US) to Taiwan and South Korea. After the correction, he is not sure whether to rebalance, diversify or exit.
How the rally began
Investor interest in AI and semiconductors accelerated in late 2022 after chip stocks bottomed following a sharp cyclical downturn. “The launch of ChatGPT in November 2022 gave the recovery a structural story because generative AI would need enormous computing power,” says Vaibhav Shah, head – products, business strategy and international business, Mirae Asset Mutual Fund.
Nvidia’s May 2023 revenue guidance exceeded expectations. The guidance showed that AI demand was translating into revenue. The trade then spread across the global AI supply chain.
Also Read
“By 2026, AI accelerator designers, custom chip and data centre networking companies, advanced chip manufacturers, memory suppliers, lithography machine suppliers, and wafer fabrication equipment providers had seen enormous valuation increases,” says Shah.
Other ancillary sectors benefited. “The rally in hardware and chip names was followed by rallies in power infrastructure, utilities and cloud providers,” says Sumanta Khan, fund manager – equity, Edelweiss Asset Management.
The correction
The Philadelphia Semiconductor Index (SOX) and South Korea’s Korea Composite Stock Price Index (Kospi) peaked on June 22. The correction began in late June 2026. Valuations got sharply repriced in July. Fears about memory-chip demand drove Kospi’s sell-off in July.
The second leg of the correction began in October, deepened subsequently by the US midterm results. US, Korean and Taiwanese markets fell sharply on November 5 and 6.
What drove the sell-off
Investors sitting on large profits booked gains. Investors have also begun to question whether the capital being spent on building the AI infrastructure is excessive. “J.P. Morgan estimated that capital spending would absorb 90 to 100 per cent of the largest cloud companies’ operating cash flow in 2026,” says Subho Moulik, founder and chief executive officer (CEO), Appreciate.
“They also question whether the promised productivity gains would be achieved,” says Khan.
AI infrastructure increasingly depends on debt, but lenders demand higher yields or step back. “Debt funding raised concerns about whether infrastructure projects would generate enough cash to service repayments,” says Khan.
The 10-year Treasury yield crossed 5 per cent intraday on November 10. This put pressure on growth stocks with high multiples. “Rising interest rates, amid a hawkish outlook from global central banks, were expected to further suppress returns on investment,” says Khan.
Democrats’ midterm gains raised doubts about new data centres. “Markets worried that stricter rules on data-centre siting, permitting and power following the US midterm results could slow or delay the AI build-out regardless of underlying demand,” says Shah.
Data centres are likely to pay more for electricity. “Higher power bills lower the return on each new data centre, which determines how much cloud companies spend on chips,” says Moulik.
Concentration risk
A few stocks dominate the Kospi, the Taiwan Stock Exchange Capitalisation Weighted Stock Index (Taiex), and the SOX, allowing stress in these companies to spread across an index. “Holding US, Korean and Taiwanese equities means holding the same AI trade three times,” says Moulik.
Investors exposed to one sector become highly dependent on its performance.
Execution challenges threaten the AI theme. It remains unclear whether current investments are justified. “Enterprise adoption of AI remains a key risk. It remains to be seen when adoption will pick up as use cases grow,” says Khan.
Where investors went wrong
Strong returns drew many investors into portfolios heavily tilted towards AI and semiconductor stocks and the US, Taiwan and Korea. “Fear of missing out prompted them to invest more than originally intended at a time when India underperformed international markets,” says Vishal Dhawan, founder and CEO, Plan Ahead Wealth Advisors.
Another mistake was buying the theme after a spectacular run-up in tech stocks. “LRS investments in overseas equity and debt more than doubled to $321.5 million in August 2026, just as AI and chip stocks reached peak valuations,” says Manish Bhandari, founder, CEO & portfolio manager, Vallum Capital. He adds that Indian investors were effectively part of the retail crowd that gave institutions an exit near the top in the global markets.
The paucity of options led many investors to buy internationally focused ETFs at significant premiums to their net asset values in the Indian market.
Rebalance exposure
First, decide your ideal allocation to international equities. “If your original allocation was 20 per cent of the equity portfolio, you need to rebalance back to that level if you have become overweight,” says Dhawan.
Check whether your foreign portfolio is excessively concentrated. “Portfolios concentrated in AI and semiconductor stocks or ETFs, technology or the Magnificent Seven need to diversify across sectors,” says Dhawan.
“Thematic semiconductor or AI funds should be satellite holdings, ideally no more than a fifth of the foreign allocation,” says Bhandari. He suggests consolidating into one broad, low-cost core holding, such as an S&P 500 or global index fund, rather than three funds owning the same AI names.
Investors who borrowed to join the AI rally should unwind leverage. Those with exposure through an employee stock option plan (ESOP) should also diversify beyond that stock.
Check stock quality
Direct stock investors should distinguish companies with strong balance sheets from highly leveraged ones. “Avoid averaging down without assessing the stock’s quality,” says Dhawan.
Prefer a diversified semiconductor ETF over a handful of stocks unless you research companies deeply. Also, avoid using margin, options or leveraged ETFs to recover faster.
Avoid these traps
Avoid exiting all international exposure in response to the correction. International equities remain a core holding with a key role in the portfolio.
Limit AI exposure, but do not exit it completely. “The core AI demand story remains intact,” says Shah.
“Fix a foreign allocation in advance, typically 10–25 per cent of equity, and rebalance periodically rather than chasing headlines,” says Bhandari.
Avoid chasing momentum or acting out of fear of missing out (FOMO) when prices rise rapidly. Such behaviour led to over-concentration in the AI theme. Finally, have a long-term horizon when investing abroad, as you do domestically, and eschew short-term trading.
More From This Section
Topics : Artificial intelligence semiconductor Nvidia
Don't miss the most important news and views of the day. Get them on our Telegram channel
First Published: Nov 12 2026 | 10:02 PM IST
