We’ve entered the golden age of access.
In minutes, you can surface earnings call summaries, industry trends, governance history, even forensic flags. A junior analyst in New Delhi today has more research firepower than a New York based hedge fund portfolio manager in 2005.
AI gives us speed, structure, and conviction. What it also gives us—quietly—is the illusion of mastery.
And that is dangerous.
⚠️ Intelligence ≠ Judgment
Ask a chatbot why a stock is attractive, and it will give you 10 bullet points. Throw in sectoral tailwinds, compare valuation multiples, add a quote from management.
You walk away not just informed—but convinced.
But as Daniel Kahneman warned, confidence is a poor proxy for accuracy. The more coherent a story sounds, the more likely we are to act on it—even if it’s built on sand.
AI may hallucinate facts. It definitely conjures conviction.
📉 LTCM: The Smartest Guys in the Room Still Drown
Let’s not forget Long-Term Capital Management (LTCM). Run by Nobel laureates and battle-tested traders, it collapsed in 1998 after placing massive, leveraged bets on seemingly safe arbitrage trades.
Their assumptions weren’t irrational. The math made sense — under normal conditions.
But investing isn’t done in a lab. When Russia defaulted, markets panicked, liquidity disappeared, and the trades stopped converging. What followed wasn’t a flaw in the formula — it was a brutal lesson in tail risk and fragility.
AI thrives in deterministic worlds — physics, math, chess. Domains with clean inputs and defined boundaries. It outsmarts grandmasters where the rules stay constant.
But markets? They’re reflexive, messy, human. Probabilistic, not predictable.
And in this game, the smartest minds can stumble – or go bust. LTCM. Archegos. Tiger. Even sovereign giants like GIC, who posted just 3.8% CAGR over two decades.
So, when AI gives you conviction, remember:
You might be the college swim captain — but that doesn’t mean you can survive the open sea.
🧠 More Knowledge, Same Behavior
Barber & Odean (2000) found that overconfident investors trade more — and earn less.
That was pre–Robinhood, Reels, Reddit.
Since then, we’ve had:
- Free stock tips
- Discount brokers
- Finfluencers
- Screener tools
- Now, AI that gives you 10 ideas before your chai cools
Still, SEBI’s 2023 study showed that 89% of retail F&O traders lost money in FY22 — average loss ₹1.1 lakh. And this during a raging bull run.
Globally, DALBAR’s 20-year data shows retail investors underperformed the S&P by 3%+ annually — just from poor timing and bad behavior.
We’ve made information free.
But temperament? That’s still earned the hard way.
📍 Everyone Has the Same Map
You ask AI for “undervalued compounders under ₹5,000 Cr market cap.”
So does everyone else.
And now you’ve got thousands of smart-sounding people crowding into the same trade, chasing the same setup, using the same language.
What happens when the story changes?
Nothing breaks consensus faster than a mild earnings miss.
The insight isn’t wrong. The illusion of exclusivity is.
🎭 The Rise of the Convincing Salesman
Earlier, most advisors were just salesmen with a product list. “ELSS for tax,” “balanced advantage for safety” — standard pitch.
Now with AI, they’ve levelled up. They talk like strategists — throw around words like “volatility drag,” “credit spreads,” “duration play.”
But jargon doesn’t equal judgment.
You wouldn’t let someone with a few Coursera videos do surgery on you.
But you’ll let them manage your life savings?
AI lets everyone sound smarter. Doesn’t mean they are.
🛑 Re-anchoring: How to Actually Use AI (Without Drowning in It)
This isn’t a call to reject AI.
It’s a call to use it better.
Let AI be your research analyst.
Not your investment committee.
Use it to:
- Scan quickly across sectors and screens
- Catch red flags (pledges, auditors, anomalies)
- Summarize filings you’d otherwise never read
- Test narratives — ask it for the bear case
- Compare patterns — spot disconnects others miss
But don’t stop there.
Before hitting “Buy,” run a 30‑second firewall:
| 🔍 Ask This | ✅ Why It Matters |
|---|---|
| What’s AI not seeing? | Tail risks, regime shifts, promoter behavior |
| Am I just persuaded by coherence? | Slick stories hide weak probabilities |
| Who else is in this trade? | Crowding = faster exits, harsher drawdowns |
| Am I inside my circle of competence? | Knowing the jargon ≠ understanding the business |
| What’s the real downside? | Position sizing isn’t sexy — it’s survival |
📜 Timeless Rules Still Apply
Let’s bring back what AI won’t remind you of — unless you prompt it very specifically:
- Circle of competence
- Margin of safety
- Process over prediction
- Avoiding ruin > chasing returns
- Temperament > talent
Investing is not a quiz or an argument to win. It’s not about having the smartest sounding take.
It’s about beating the indices. Consistently.
💡 TL;DR for the AI-Age Investor
| Principle | Why it Matters |
| Let AI sweep the room, not call the shot | Use it to scan, not decide |
| Suspect perfect stories | Slick narratives = hidden risks |
| Dodge the herd | The crowd is now faster than ever |
| Anchor your brain | Re-read Kahneman or Howard Marks |
| Run the playbook, not the hotlist | Process compounds. Tips expire. |
AI is a brilliant intern, not a portfolio manager.
Keep the toolkit—lose the god complex.
