How To Improve Your Trading With ChatGPT – Pt 1
Larry Connors’ Trading Lesson of The Day | February 19, 2025
AI trading, especially with ChatGPT, is the next generation of trading.
An experienced trader who also knows how to use ChatGPT can literally now accomplish in 20 minutes what it took a team of 25-40 professionals at a multibillion-dollar hedge fund 4-6 weeks to do only 4 years ago!
AI Trading is here to stay, and those who are properly using it today are already light years ahead of other traders.
In this multi-part trading lesson series, I’m going teach you how to put in a framework for yourself to be able to create new trading strategies, optimize your trading, etc., etc., etc.
In order to professionally do this, remember: Everything is based on how you prompt!
The better you prompt, the greater the output will be from ChatGPT and every other Large Language Model like Claude, Gemini, Grok, etc.
In late 2023, I taught the first class on Wall Street, teaching professional traders how to use AI to improve their trading. That course has now been advanced, teaching professional traders how to build high-performing hedge fund-level trading strategies.
The feedback and the trading abilities of many of these professionals have been astounding.
My goal here in the multi-part series is to make sure you have the best foundation in place to improve your trading with AI. This means learning the most important thing – how to tap into the vast amount of knowledge inside AI, especially ChatGPT.
We’ll do this one step at a time, starting right from the beginning.
Whether you were fortunate to have been there in Dec 2022 with ChatGPT as I was, or are using it for the first time, there’s something here for everyone.
This foundation is a must to optimize AI. It’s also the same foundation I teach and also use for my personal trading.
You’ll learn not only how-to prompt, you’ll also learn the “why” behind the how-to.
This is the equivalent of me teaching you how to fish for life instead of me handing you a fish to eat one time.
The first is lifelong knowledge that you will be able to apply to continuously improve your trading. And as GPT and other LLMs rapidly advance their capabilities, everything you learn here will be further boosted by these advancements.
This is a revolutionary once-in-a-generation opportunity for traders to take advantage of. What you’ll see in this series will likely blow you away.
And it’s why, after being on a Bloomberg terminal since 1989, I cancelled my $30,000 subscription. In my opinion, as of today, Bloomberg is last generation’s technology. It’s unable to do a fraction of what ChatGPT can do…it’s not even close!
If you’re professionally using ChatGPT, you already know this is true. If you’re not, you’ll soon see it for yourself.
Sign up below if you’d like to attend a free webinar I’m teaching on Tuesday, February 25:
Lesson One
ChatGPT Doesn’t Know Who It’s Supposed To Be
Amateur (mainstream) prompters think they can just bluntly type any question into ChatGPT and it will immediately give them the very best answer.
It’s a mistake the majority of the world makes…then they proclaim “AI doesn’t work” or some other uninformed excuse.
In this series, I’m going to show you how it really works in order for you to get the maximum knowledge from ChatGPT.
I’m assuming you’re smart. Likely very smart. With your intelligence, you can answer a lot of questions about a lot of things.
Now, how would you feel if a total stranger on the street put their face in front of you and started asking you random blunt questions?
How do I bake a cake?
My stomach hurts. What do you think I have?
Tell me about the history of the Civil War.
How do I do my taxes?
What stocks are going to appreciate at least 100% tomorrow?
Etc., etc., etc.
How would you respond? A stranger abruptly starts peppering you with random questions on random topics absolutely out of nowhere.
They know nothing about you, nor you about them. All they know is that you’re smart, and of course, smart people know everything!
Your first reaction would be…WTF?
Well, ChatGPT doesn’t say WTF (though it should).
It gives a “very basic pedestrian answer” to the question.
That’s it!
Why?
Because… ChatGPT (and all AI) doesn’t know who it’s supposed to be. 👉 You have to tell it who it is!
I learned this early on, and it was further reinforced to me by Caleb Briggs, who consults for me.
Caleb is the lead author of the “AI Conundrum” published in 2024 by MIT Press.
Forbes recently reviewed it and said it was one of the best business management books about AI it’s read in a few years:
…Overall, this is one of the best management books about AI I’ve read in a few years. I strongly recommend it to those people trying to better understand how Al works, the risks and rewards for business based on the current state of the art, case studies showing actual implementations, and a nice framework for business evaluation of AI use. -David A. Teich | Forbes
In one of our conversations, Caleb clearly explained to me why it’s so important to make sure you get AI (ChatGPT) to know who it is.
Because once it knows who it is, it becomes it.
In trading, ask it a basic trading question, and you’ll get a basic trading answer.
On either hand, tell it it’s the head of a multibillion-dollar hedge fund, or it’s an expert whose clients include hedge fund managers like Ken Griffin, Steve Cohen, and many of the elite other hedge fund managers in the world, and the depth of its response moves up into a whole new level.
Here’s an example of a prompt I wrote to show you;
Prompt 1
“What is an ETF?”
ChatGPT:
An ETF (Exchange-Traded Fund) is a type of investment fund that holds a diversified portfolio of assets—such as stocks, bonds, commodities, or other securities—and trades on an exchange like a stock.
Key Features:
1. Diversification – ETFs often track an index (e.g., S&P 500, Nasdaq-100) or a specific sector, providing broad exposure to multiple assets.
2. Liquidity – Unlike mutual funds, which only trade at the end of the day, ETFs can be bought and sold throughout market hours at market prices.
3. Lower Costs – Most ETFs have lower expense ratios compared to actively managed funds since many track an index passively.
4. Flexibility – Investors can trade ETFs like stocks, using strategies such as stop-loss orders, options, and short selling.
5. Tax Efficiency – ETFs are structured to minimize capital gains distributions, making them more tax-efficient than mutual funds.
ETFs are widely used for various strategies, including passive investing, thematic investing, hedging, and active trading.
Examples of Popular ETFs:
• SPY (S&P 500 ETF) – Tracks the S&P 500 index.
• QQQ (Invesco QQQ) – Tracks the Nasdaq-100 index.
• ARKK (ARK Innovation ETF) – Actively managed ETF focused on disruptive technologies.
• SOXX (iShares Semiconductor ETF) – Focused on semiconductor stocks.
Would you like a deeper dive into ETF selection, strategies, or trading techniques?
The response from ChatGPT is pedestrian. Very basic. You can find this very basic explanation on Investopedia or any other mainstream site.
The follow-up questions are also very pedestrian. It’s better than the internet, but they’re very broad questions.
Prompt 2 – (Telling ChatGPT who it is)
“You’re the head of a multibillion-dollar hedge fund and an ETF expert whose clients include hedge fund managers like Ken Griffin, Steve Cohen, and many of the elite hedge fund managers in the world.
What is an ETF?”
ChatGPT said:
An ETF (Exchange-Traded Fund) is a sophisticated investment vehicle that provides liquidity, diversification, and cost efficiency while allowing for precise exposure to various asset classes.
Unlike traditional mutual funds, ETFs trade on exchanges in real-time, offering hedge funds and institutional investors tactical flexibility and execution advantages.
Institutional-Level Features of ETFs
1. Liquidity & Market Depth – High-volume ETFs, such as SPY and QQQ, offer deep liquidity with tight bid-ask spreads, making them optimal for large block trades without significant market impact.
2. Customization for Strategic Exposure – Institutional investors leverage sector-specific ETFs (e.g., SOXX for semiconductors, XLK for technology) or factor-based ETFs (momentum, value, quality) to fine-tune portfolio positioning.
3. Derivatives & Hedging Applications – ETFs are integral to options trading, delta-neutral hedging, and portfolio overlays. They allow hedge funds to efficiently express long and short views without taking on individual stock risk.
4. Tax & Cash Efficiency – ETFs’ in-kind creation and redemption process provides significant tax advantages, reducing capital gains distributions—critical for managing tax drag in large portfolios.
5. Leverage & Short Exposure – Leveraged ETFs (e.g., TQQQ for triple Nasdaq exposure) and inverse ETFs (e.g., SQQQ, SPXS) allow hedge funds to quickly scale positions or hedge systematic risk.
ETF Strategies Utilized by Elite Hedge Funds
• Macro Overlay & Risk Management – Hedge funds often use broad market ETFs to adjust beta exposure dynamically.
• Arbitrage & Liquidity Provision – Market makers and high-frequency traders engage in ETF arbitrage, exploiting NAV mispricings.
• Thematic & Quantitative Exposure – Funds deploy AI-driven models to allocate into thematic ETFs (e.g., AIQ for artificial intelligence, ARKG for genomics) based on machine learning-driven factor rotations.
• Event-Driven & Catalyst Trading – ETFs linked to commodity cycles, rate hikes, or geopolitical trends provide a rapid, capital-efficient means to capture macro themes.
Conclusion
For hedge fund titans like Ken Griffin and Steve Cohen, ETFs are not just passive instruments but sophisticated tools for risk-adjusted alpha generation, liquidity management, and strategic asset allocation.
Would you like deeper insights into advanced ETF trading strategies, arbitrage techniques, or how leading hedge funds deploy ETFs in portfolio construction?
Exact same question. “What is an ETF?”
Yet simply telling ChatGPT who it is, a high-level ETF expert who advises hedge fund billionaires like Ken Griffin, and Steve Cohen, along with other elites in the hedge fund world, the response certainly is not pedestrian.
It’s deeper, more insightful, potentially made you smarter, showed you ways to apply it, and then it offered you the opportunity to learn hedge fund-level ETF strategies, arbitrage techniques, and hedge fund-level portfolio construction.
And with one simple enhancement – putting ChatGPT into “expert mode,” we’ve just taken a very simple question and immediately kicked the response into “professional-level trading mode.”
Also, which prompter now immediately has more knowledge to scale upon?
The one using the pedestrian prompt, which is still literally 99% of the population who is on AI (most people are not even on AI yet)?
Or the one who simply starts their first prompt by doing it professionally – telling ChatGPT who it is!
Lesson One – When you start your prompt, tell ChatGPT who it is.
It takes on that role, beginning with your first prompt and then throughout your following prompts.
This means it’s an expert right from the start. And this also means the information and knowledge you receive from it will move you ahead at a much faster pace.
Tomorrow, in Lesson Two – You’ll learn the next step. Tell ChatGPT who you are. When you do this, you’ll start receiving “customized answers” based on your expertise.
This one will take you to an even higher level…stay tuned.
Special Note – Register below if you’d like to attend a free webinar I’m teaching on Tuesday, February 25 at 1 pm ET:
👉 How To Accomplish In 20 Minutes With ChatGPT What Takes A Team Of 25-40 Professionals At A MultiBillion-Dollar Hedge Fund 4-6 Weeks To Do
Please note that Zoom handles 500 people. As with my other webinars, I expect this one to be full. If you’re interested in attending, please sign up now.
For Every Issue of Larry Connors’ Trading Lesson Of The Day, Click Here.