Widely read:
Anand Kumar/AI-Generated Image
Summary: Most readers agreed that trying to predict an AI bubble isn't the real challenge. But that only led to a harder question. If you're already investing sensibly, what are you actually supposed to do next? This story follows the conversation where the original note left off.
The responses to Dhirendra Kumar’s latest Editor’s Note, The AI bubble question, split into two clear groups. The first found the note's conclusion liberating and immediately useful and moved on. The second found it correct but incomplete, the right answer to a question that was not quite the question they were actually asking.
Both groups are worth hearing from. The second is the more interesting story.
The critique nobody held back
Harsh Daftari was direct about his disappointment. The hook was great, he wrote, but the underlying story was not worth the anticipation it created. He was not abusive about it, just honest: if you set expectations that high, you owe the reader something on the other side of them. Ntimes Wealth Creations made the same point with more specificity. The note had, in their reading, over-intellectualised the topic and left the reader with nothing they could actually use. What they wanted instead: how do successful investors of contrasting styles deal with bubble dilemmas? What is the investor's ultimate objective: to be right, or to generate a return regardless of how things develop? These are fair questions.
Amitava Tripathi reached for Shakespeare: "The fault, dear Brutus, lies not in our stars, but in ourselves that we are underlings." He meant it as an endorsement of the note's conclusion, but he added an observation that carries it further. What is remarkable, he wrote, is that in order to trap the retail investor, the promoter himself can fall into the trap of the hype he created. A bubble is not just investors catching a fever from each other. Sometimes the fever starts in the room where the product is being built.
Rajen Agrawal put all of this in a few words: "One's own itching palms and fingers are the sole cause of your wealth going kaput in your lifetime."
Where readers pushed further
The note's conclusion that a sensibly diversified investor need not worry about how the AI drama ends drew the sharpest pushback from Ramalingam. In most previous bubbles, he observed, the underlying assets retained value even after the financial meltdown. Railways lasted 50 years and gave future generations something to build on. The fibre-optic cables of the 1990s bankrupted the companies that laid them but later enabled everything that now runs on them. This time, the primary capital expenditure is the chip. Chips become obsolete in three to five years and must be replaced. The capital is gone and has to be replaced, he wrote, and it is not available like a railroad for 50 years for future growth and monetisation. This means that the trail of losses from a chip-driven bust, should one come, leads to debt holders with no future recourse and equity holders with no asset at the bottom of the rubble. He also raised the point that diversification advice works for people who can wait. It does not work for a family whose child's education falls due in the month the market hits its low and cannot wait six or thirteen months for prices to recover.
Jeslin Sugirtha pressed on a different side of the same problem. A passive investor in a broad market-cap-weighted index may appear diversified across hundreds of companies, she pointed out, but the index's recent performance and valuations may be dominated by a small group of companies tied to the same AI narrative and capital-expenditure cycle. Diversification by the number of stocks can hide concentration in a common economic factor. She asked what the note had left open: should "contained and survivable" be measured by the maximum drawdown in a single holding, by the time needed for recovery, or by the eventual impact on the family's financial goals? If a broad index fails that test, how does an investor reduce the hidden concentration without the exercise becoming an attempt to time the bubble, which is precisely what the note warned against?
Melvin offered what may be the most precise account of what a certain kind of reader actually means when they ask whether AI is a bubble. The people asking, he wrote, are not the ones who have piled into AI stocks. They are the ones who stayed away from the hype as carefully as they could. What they are not quite managing to articulate is this: when a crash comes, will it drag down the entire market and keep it depressed for years? Will it bring down financial institutions and produce a cascading effect like 2008? "The sane people are grappling with questions like that," he wrote. "The not so sane ones couldn't care less." By that reading, the note spoke to people who might be tempted to overbet, when the people actually writing in were the ones who had already stayed out and wanted to know whether that decision would protect them when everyone else's recklessness came home.
What readers are doing with it anyway
Nandkumar J found a way to hold all of this together in a cricket analogy. The market is always full of three kinds of players: the T20 player who wants to score fast and ride the momentum, the one-day player who takes calculated structural positions and accepts short-term risk and the Test player who is prepared to play in all conditions and knows exactly when to defend and when to attack. For the short-term traders and the persistently negative voices, AI will always be called a bubble. For the Test-match investor who has built a diversified portfolio, the only question that matters is whether the downside is, in the note's own phrase, contained and survivable.
Akhil S Pillai proposed the barbell approach: 90 per cent of money in ultra-safe instruments, 10 per cent in highly speculative bets where the maximum loss is capped. He was practical about the limits of this in the Indian context. There is no straightforward AI fund or ETF available for the retail investor. The RBI's $7 billion cap on overseas fund investments, set in 2008, has not been raised in 18 years. The LRS route that remains open is effectively only for high-net-worth investors. The strategy makes sense. The instruments to execute it cleanly are not yet there.
Subramanian Venkateshwaran, who follows Ruchir Sharma's market predictions closely, summarised Sharma's diagnosis of the current AI moment: over-investment, over-leverage, over-ownership, over-valuation. He added the Technology S Curve and the Gartner Hype Cycle as frameworks he has tracked across multiple technology generations, from mainframes through client-server to internet to social-mobile-analytics-cloud, each cycle shorter than the last. His concern is not whether the technology is real but how fast the hype travels now. Information moves so fast and social media algorithms amplify it so effectively that the crowd forms and panics faster than any previous generation of investors has experienced.
The note's closing line, that the bubble in you is the only part of the entire affair you control, and happily also the only part that matters, landed well with most readers. It is clean, it is true and it is genuinely calming. The harder replies were not saying it was wrong. They were saying: yes, and now what? How do we apply the principle when the index itself may be the hidden concentration risk, when the assets underlying the crash depreciate in years rather than persisting for decades and when the people most likely to be hurt are not the overenthusiastic speculators but the carefully diversified families whose timelines cannot wait for a market to find its floor?
Those are not comfortable questions. But they are the ones this week's inbox was actually asking.
Credits
Harsh Daftari, Ntimes Wealth Creations, Amitava Tripathi, Rajen Agrawal, Ramalingam, Jeslin Sugirtha, Melvin, Nandkumar J, Akhil S Pillai, Subramanian Venkateshwaran
Also read: Save more, invest better: The elders already knew it
This article was originally published on July 22, 2026.
Ask Value Research 
New episodes every Friday 
New Process Alert 
Seems like you've uploaded a wrong QR code. Please upload the QR code downloaded from MF Central platform.
