Daybreak - LLMs are the new pied pipers of the stock market

Episode Date: August 16, 2026

An HSBC survey found that 86% of affluent Indian investors are now using AI in their trading — the highest share of any country surveyed, above China, Singapore, and the US. Nearly half say AI tools are their primary source of investment ideas.Most of them are using the same two or three chatbots. And a growing body of academic research has found that when millions of investors ask the same models the same questions, they get the same answers — and pile into the same stocks at the same time.Goldman Sachs has already started modelling it. They can see the trades coming.Daybreak is produced from the newsroom of The Ken, India’s first subscriber-only business news platform. Subscribe for more exclusive, deeply-reported, and analytical business stories.

If you ever asked AI to pick your next stock investment, you might want to think again. Let me explain why. This June, HSBC, released a survey of 1,000 rich Indians in which it found that more than three-fourths of them actively use AI for their stock market investments. Out of them, more than 40% even went on to admit that the AI tools they were using were the main source of their ideas. But think about which AI chatbot. is always open on your laptop.

Chances are, it's most likely either OpenEI's chat GPT, Anthropics Claude or Google's Gemini. Which raises the question, my colleague Mutasim Khan, asks in his piece. If everyone is using the same three chatbots for their stock picks, what goes wrong? To answer that question, I'm going to be reading out an edition from one of the Ken's most popular subscriber-only newsletter called Kaching. And it's titled,

LLMs are the new pipe pipers of the stock market. Welcome to Daybreak, a business podcast from the Ken. I'm your host, Trey Triergergeese, and every day of the week, my co-host, Nicka Shama and I will bring you one news story that is worth understanding and worth your time. Today is Monday, the 17th of the forest. Spent a summer as an intern on a trading desk once, and realized fairly quickly that I was not very good at it.

This was long enough ago that good old public forums like Stack Overflow and R slash Learn Python was still the go-to over chat GPD for coding advice. The fun part of my job was making wild hypotheses about the market and hoping they work. All the rest, like writing code, fetching data, learning new strategies, made me yawn around a little too much for my manager's liking. But the same friction that deterred me from turning an idea into an actual trade is also the friction large language models or LLMs have been easing remarkably fast ever since. An HSBC survey of about 1,000 affluent Indian investors released in June found that 86% of the respondents are now using AI in their trading flows,

which is the highest share of any country the survey covered, about China, Singapore and the US. 42% of them counted AI-powered tools among their leading sources of investment ideas, second only to China's 48%. But individual investors are not the only invitees to the LLM gala. Finfluensers or social media content creators who share financial advice are stacking chat GPT prompt packs on tops of the tips, courses and telegram boards that they are already selling. The internet is also flooded with guides like,

50 chat GPT prompts for intraday and short-term traders, and entire prompt libraries, almost all of them opening with some variant of act like you are a professional Indian market analyst. And if prompts feel like too much effort, GitHub, which is Microsoft's developer platform, is full of open-source trading boards that plug an LLM into your brokerage account,

allowing a simple chat with, say, Claude to drive your trades. It appears that LLMs are becoming more and more embedded into markets from every direction. The premise behind this whole movement is essentially that AI is on its way to leveling the playing field between individual and institutional traders and that early movers get a short but glorious window of easy money before the rest of the market catches up and closes it. But still, there is a catch. LLM usage is almost entirely concentrated within three models, OpenAI's chart GPT, Google's Gemini and Anthropics Clod. And a growing body of academic work has begun documenting something specific

about what these models tend to do at scale. LLMs, by design, are non-deterministic. No two answers are ever identical even to the same prompt. But run the same kind of query enough times across a large enough sample and a distribution emerges. When it comes to financial markets, LLMs have been shown to have systemic biases in that distribution. Biasis not toward any particular stock per se,

but towards stocks with certain characteristics. Disque, for example, toward large-cap stocks over small ones, toward technology over other sectors, toward contrarian setups over momentum, and toward buying rather than selling. The paper argues that since major LLMs are trained on broadly the same universe of public financial writing, they end up sharing the same systematic biases in the aggregate. Which is a strange kind of level playing field.

The retail trader gets tools that in the aggregate make her more predictable to the very institutions she thought she was catching up to. In a June podcast, Osman Ali, the global co-head of Goldman Sachs quantitative, investing arm, put it plainly. He said, if you ask the same model, the same kind of question, you get the same type of answer, which means investors pile into the same type of securities at roughly the same time and prices get pushed away from any sort of fundamental value. Whether or not retail investors were consciously using LLMs to make their decisions, Ali said that the crowding effect was already visible. Broad LLM use had produced a different type of predictability and inefficiency in the market.

Millions of retail investors using essentially the same two chatpots may not necessarily democratize markets. Oswald claimed that this also makes retail behavior highly predictable. His team, he added, had begun spending a lot of time modeling how retail was using AI so that they could actually see the trades coming before retail investors placed them. An academic paper from March takes an especially ingenious approach to this question as well.

Showshwe-Hur wanted to understand what chat GPT was doing to the US markets and realized she could study it by studying retail trader behavior every time the chatbot crashed. Each crash was, in effect, a natural experiment. If chat GPT was really substantially shaping how retail investors traded,

there would be a discernible change in market behavior. when retail traders lose access to the LLM. And there was. Retail trading volume dropped by about 6% every time ChargapD crashed, which was itself striking because it meant that a meaningful share of American retail activity was, in some way, being routed through a single LLM. Markets in principle work because different people show up with different views. Some see a stock as undervalued, some as overvalued.

some as overvalued. Some sell into a rally, some buy into a fall, and it is precisely this disagreement that keeps prices honest. Shau Chau Chay found that when Chatt GPD was on, that disagreement tended to thin out. Retail investors began interpreting the same stocks the same way, and small, useful frictions that let a market discover a fair price began to erode. But what about in Indian markets?

A trader at an Indian proprietary firm told me his desk had been running strategies built around herding behavior in retail investors and those strategies had started performing meaningfully better in the last few months. He said that there was no clean way to prove what was causing it, but a good guess was that it came down to LLM-assisted trading. In fact, Seby II in its consultation paper on the responsible usage of AI and machine learning in Indian securities markets, explicitly flagged the risk of herding and collusive behavior arising from widespread use of common models and datasets by large cohorts of market participants. Be that as it may, this argument only holds if Chad GPT is the one doing the actual thinking. And that is not necessarily the case.

LLMs have also powered a whole new class of infrastructure builders on the ground. Rajandran R is one of them. A long-time systematic trader, he built Open Algo, a self-hostable WordPress-style platform that lets a non-programmer build and deploy a trading strategy without writing a single line of code. Rajantran has watched this shift happen room by room. He shared an experience where two years ago, in a session of 150 people, he asked how many of them used chat GPT. Hardly four or five hands went up then. But at an algorithm trading meet he was hosting recently, where roughly 400 people between the ages of 25 and 70 had shown up, he said almost 80% of them were already using Chad GPD or Claude to help with their investing journey. However, from Rajan Trin's vantage, Indian retail is nowhere near the point where the LLM convergence effect would begin showing up in the market.

Most of the retail traders he sees use LLMs for execution, not for idea generation. He said it looks more like someone having a design in mind and using AI to convert that design into code. However, the trader at the proprietary firm also offered a warning. He said it is only by chance that an LLM will actually come up with a profitable trading strategy all by itself. If LLMs can find a piece of information, the market has most likely already priced it instantaneously and any profitable signal has already decayed. Still, in the rare case your LLM buddy does hand you a profit-making formula, don't tell anyone. As an ex-trader friend once told me, no one who makes money opens their mouth and no one who

opens their mouth actually makes money. These are words to trade by. Daybreak is produced from the newsroom of the Ken India's first subscriber-focused business news platform. What you're listening to is just a small sample of our subscriber-only offerings. A full subscription offers daily long-form feature stories, newsletters and a whole bunch of premium podcasts. To subscribe, head to the ken.com and click on the red subscribe button on the top of the Ken website. Today's episode was hosted and produced by my colleague Rachel Virgis and edited by Rajiv Sien.