Why ChatGPT Will Ruin Your Portfolio (And How Specialized AI Finds Multibaggers Until 2034)

A Personal Learning Log

In October 2026, I attended an online training session on AI-driven multibagger investing conducted by Akshay Gulati (an IIT Guwahati alumnus and active market investor since 2018), hosted in partnership with The Economic Times. The session was designed as a roadmap for anyone looking to leverage artificial intelligence to build wealth between now and 2034. I am publishing this post as a permanent digital record of my key takeaways, notes, and personal verdict on the tools and frameworks shared during the training.

TL;DR Version

• General AI fails at finance. Tools like ChatGPT and Gemini lack access to terabytes of private Indian market data, suffer from sycophancy bias, and can or will hallucinate financial metrics .
  

• Retail investors lose due to five hurdles. Lack of time, sector knowledge, financial fluency, entry/exit timing, and "borrowed conviction" keep everyday investors stuck buying at the top.
  

• Great investing balances three core pillars. Every legendary investor—from Warren Buffett (valuation) to Peter Lynch (growth) and Charlie Munger (quality)—simply orders these three variables differently.
  

• The SCRAP Framework filters out noise. By screening for 15%+ sector growth, 25%+ company compounding, strict red flags, smart allocation, and emotional discipline, we can systematically hunt for multibaggers.
  

• Purpose-built AI changes the game—with caveats. Platforms like SuperInvesting AI automate fundamental formulas across 6,000+ stocks, though DIY investors can still extract massive value from foundational tools like Screener.in.

Lesson 1: The Rearview Mirror Trap & The EV Case Study

One of the biggest takeaways from Akshay’s session is a simple truth: investing always looks easy in the rearview mirror. Looking at a five-year chart, anyone can point to a tenbagger and claim it was obvious. In reality, you cannot make money on borrowed conviction. Relying on stock tips from friends, TV pundits, or social media guarantees you will panic-sell at the first sign of volatility.

To prove how structured AI beats surface-level human intuition, Akshay walked us through a live backtesting challenge using three Electric Vehicle (EV) companies:

• The Human Trap: Most of us looking at surface-level financials and recent stock momentum would naturally pick the first two EV companies, which looked like clear winners on paper.

• The AI Advantage: The AI model rejected the top two favorites and picked the third company. By parsing deep order-book disclosures and vendor filings, the AI spotted that the first two companies were walking straight into severe supply chain bottlenecks, whereas the third company had secured an incoming order pipeline shielded from those disruptions.

Historical market data proved the AI right. That case study made it clear to me why everyday investors fall behind—and how data-backed ideas remove guesswork.

Lesson 2: The 5 Hurdles Holding Retail Investors Back

Before diving into AI tools, Akshay broke down the five structural roadblocks that prevent retail investors from compounding wealth:

1. The Language of Finance: Balance sheets, cash flow statements, and ratios like ROCE or EBITDA feel like a foreign language when you are starting out.
   

2. Lack of Sector Knowledge: Understanding how macro tailwinds affect specific industries (like defense or pharmaceuticals) takes years of study.
   

3. Time Constraints: There are roughly 6,000 publicly listed companies in India. Nobody with a full-time job can manually read annual reports and earnings transcripts for thousands of stocks.
   

4. Weak Conviction: Without doing your own deep research, every 10% market dip feels like a crisis, forcing you to exit great companies too early.
   

5. Poor Entry and Exit Timing: Retail investors routinely buy when hype is at its peak and sell when fear takes over.
   

When you remove these five hurdles with a clear, data-backed thesis, everything changes. However, my biggest wake-up call from the session was realizing why standard AI chatbots cannot solve this for us.

Lesson 3: Why ChatGPT and Gemini Fail at Stock Research

If you open a standard AI chatbot and ask for the best Indian stocks to buy, you are setting yourself up for costly mistakes. Here is what I learned about why general-purpose Large Language Models (LLMs) fail at investing:

• Built for Conversation, Not Finance: General models like ChatGPT and Gemini are trained to give pleasant, well-structured answers to broad questions. Just as models like Claude are heavily optimized for coding, stock research requires an engine specifically fine-tuned for financial math and corporate filings.
  

• The Data Wall: Analyzing ~6,000 listed Indian companies requires processing terabytes of structured financial filings, shareholding patterns, and historical disclosures—much of which is not cleanly accessible to public chatbots.
  

• Dangerous Hallucinations: Because general chatbots lack live, verified database access, they guess. If you ask ChatGPT for HDFC Bank’s Return on Equity (ROE) in 2023, it may confidently invent a plausible-sounding figure. Making real investment decisions on hallucinated numbers is lethal.
  

• Sycophancy and Popularity Bias: General AI is designed to please the user and mirror internet consensus. If 20 users ask about ICICI Bank and only 10 ask about HDFC Bank, a general LLM will naturally lean toward talking more about ICICI simply because it has more conversational weight behind it.
  

To solve this, Akshay and his team have worked since 2024 on building an India-specific financial LLM (backed by a published whitepaper) enriched with verified market data stretching back to 2015. Having backtested this model across market cycles, Akshay predicts that within the next year, specialized AI will handle nearly all stock research and screening, leaving humans to focus strictly on judgment and capital allocation.

Lesson 4: The 3 Pillars of Every Legendary Investor

No matter how smart AI becomes, it must still follow proven investing principles. A core concept I noted from the training is that every great investor plays around three attributes: Quality, Growth, and Valuation.

What separates investing legends is simply the priority order in which they rank those three variables:

• Valuation First (Warren Buffett & Benjamin Graham): Classic value investors demand a steep margin of safety, refusing to overpay no matter how exciting the story sounds.
  

• Quality First (Charlie Munger): Buffett’s longtime partner famously shifted their philosophy toward buying "wonderful businesses at fair prices" rather than mediocre businesses at cheap prices, prioritizing strong competitive moats and high return on capital.
  

• Growth First (Peter Lynch): The legendary Fidelity Magellan Fund manager focused on fast-growing companies ("tenbaggers") with strong earnings momentum before institutional investors caught on.

Lesson 5: The SCRAP Framework for Hunting Multibaggers

To turn these principles into a repeatable checklist, Akshay taught us the SCRAP Framework (inspired by the philosophies of Peter Lynch and Rakesh Jhunjhunwala):

1. S – Sector Analysis (The 15% Rule): Never swim against the tide. Look for sectors growing faster than India's GDP—specifically industries expanding at 15%+ CAGR. Defense and pharma had massive runs recently, but investing is about the future, not the past. Once you identify the right tailwind sector, half your work is done.
   

2. C – Compounding Machines (25% YoY): Within that high-growth sector, isolate businesses capable of compounding earnings at 25% year-over-year, backed by strong ROE, ROCE, and healthy EBITDA margins.
   

3. R – Rejection Criteria (Red Flags): Great investing is largely about avoiding losers. Immediately filter out companies with high debt, poor governance, promoter share pledging, or weak cash flows.
   

4. A – Allocation & Portfolio Management: Decide how much capital goes into each high-conviction idea, and set clear rules for staggered entry and disciplined exits.
   

5. P – Psychology & Inflection Points: The market transfers money from the impatient to the patient. You must overcome shortsightedness, hold through volatility with genuine mental conviction, and re-evaluate your thesis every few years to check if the business has hit a major inflection point.

My Verdict: SuperInvesting AI vs. Screener.in

Manual calculation of CAGR, EBITDA growth, ROE, and future growth across thousands of stocks is exhausting. To automate the SCRAP framework, Akshay showcased SuperInvesting AI—a platform that runs automated fundamental formulas and includes AI Investor Personas modeled after real-life legends like Warren Buffett and Peter Lynch to suggest and critique stock ideas.

My Honest Take After Digging Deeper:
Looking closely at the product, SuperInvesting AI appears to be a smart, attractive package—especially if you lock in their 3-year subscription bundle. However, doing a bit of my own homework outside the session, it is clear that the platform builds upon the same foundational financial data ecosystem popularized by Screener.in.

• Where SuperInvesting AI shines: It layers conversational AI, automated SCRAP scoring, and investor-persona bots directly on top of raw financial data. If you want speed, guided synthesis, and instant red-flag checks without building Excel models, it is a compelling co-pilot.

• Where Screener.in holds its ground: Screener.in remains an unbeatable powerhouse for hands-on investors. It offers generous free access, customizable ratio queries, and clean financial statements at a much lower subscription cost with plenty of built-in value.

Final Summary

Documenting this October 2026 training reinforced one major lesson for my own investing journey through 2034: blindly trusting general AI chatbots like ChatGPT for stock picks is a recipe for disaster. Whether you decide to pay for an AI-layered co-pilot like SuperInvesting AI or stick to running custom screens on Screener.in, the core edge remains yours to execute—pick sectors growing faster than GDP, demand 25% compounding, ruthlessly eliminate red flags, and build enough conviction to hold through the noise.

Over to you: Have you tried using AI tools for your stock research yet, or do you still rely on Screener.in and manual annual report reading? Share your experience in the comments or email me.

Thank you for reading this week's analysis. For editorial feedback, technical corrections, or submissions, please contact us at: reach@anshumaanbakshi.com.

2. Matching LinkedIn Post

Stop asking ChatGPT which stocks to buy. 🛑

In October 2026, I attended an eye-opening training session on AI-driven multibagger investing led by Akshay Gulati (in partnership with The Economic Times). I put together a complete digital record of my learnings and personal verdict on the tools shared—here are the biggest takeaways:

• General AI fails at finance. Tools like ChatGPT and Gemini lack access to terabytes of private Indian market data, suffer from sycophancy bias, and will hallucinate financial metrics like HDFC Bank's ROE.

• Retail investors lose due to five hurdles. Lack of time, sector knowledge, financial fluency, entry/exit timing, and "borrowed conviction" keep everyday investors stuck buying at the top.

• Great investing balances three core pillars. Every legendary investor—from Warren Buffett (valuation) to Peter Lynch (growth) and Charlie Munger (quality)—simply orders these three variables differently.

• The SCRAP Framework filters out noise. By screening for 15%+ sector growth, 25%+ company compounding, strict red flags, smart allocation, and emotional discipline, we can systematically hunt for multibaggers.

• Purpose-built AI changes the game—with caveats. Platforms like SuperInvesting AI automate fundamental formulas across 6,000+ stocks, though DIY investors can still extract massive value from foundational tools like Screener.in.

Want to read my full notes on the EV stock case study, the SCRAP checklist, and my honest verdict comparing SuperInvesting AI vs. Screener.in?

Read the full breakdown here 👉 [Insert Blog Link Here]

#StockMarketIndia #ValueInvesting #AIinFinance #MultibaggerStocks #WealthCreation

Connect

Explore my services and portfolio for growth.

Inspire

Create

reach@anshumaanbakshi.com

+91 78278 45113

© 2026. All rights reserved.

Like this website ?? Own a similar one! Click here to learn more