Boost your learning journey with GPT-4o

I’ve always been convinced people can achieve everything with the right mindset. But it takes effort and a few mental tweaks.

Fortunately, the new Large Language Model capabilities can make learning easier, faster, and much more engaging!

What it takes to learn efficiently

Swedish psychologist Anders Ericsson, the inventor of “the 10,000-hour rule“ said:

It takes 10,000 hours of intensive practice to achieve mastery of complex skills and materials

Ericsson estimated that expert performers get focused sessions of 3–4 hours of work to stay ahead of the game.

10,000 hours rule

10,000 hours is massive. That’s right. 10,000 hours in a lifetime is a LOT, and you’d be allowed to feel discouraged, feeling that your current efforts at learning a new skill are just a drop in the water.

Know that Ericsson’s theory has since been proved wrong, by the University of Princeton and some other studies.

But even if you don’t aim for mastery (not everyone wants — and can — be the next Mozart, or Ronaldo), the idea is still relevant.

You’ve got to put in the work. The right kind of work. One that fits your way of learning, sparks your curiosity and fuels your motivation.

1 hour rule

Earlier in history, in the 1950s, the American Earl Nightingale studied the concept of “deliberate practice”. His book, ‘The Strangest Secret” was released in 1956 and contains a now famous quote:

One hour per day of study in your chosen field is all it takes. One hour per day of study will put you at the top of your field within three years. Within five years you’ll be a national authority. In seven years, you can be one of the best people in the world at what you do — Earl Nightingale

One hour of study. Reading non-fiction books, for instance.

Brian Tracy is a well-known self-development author and speaker. His talks and teachings insist on one thing:

If you read a book in your chosen field for one hour a day, that would translate to 1 book per week. 1 book per week translates to about 50 books per year — Brian Tracy

Work compounds and counts towards your knowledge and expertise, as interests do with your wealth.

Seems way more achievable, right? Feeling motivated yet? Wait until the next part.

Reading books is great. There is no debate here. But let’s put it that way. To get the best out of my 50 books a year, I would have to:

  • Make a list of books in my field of interest
  • Read reviews, to see if they’re worth reading (not all books are made equal)
  • Buy the paperbacks, or ebooks (or find them in the library)
  • Study the excerpts to set a logical and pedagogical order of reading them. You don’t want to learn how to jump hurdles before you know how to walk
  • 1 hour goes by fast. It would be counter-productive to waste half of it reading introductions, ToC, or chapters that are not directly related to the goal you set.

Supercharge your learning

For a while now, AI assistants have been around us, and many LLM models are available, for free or not, as SaaS or not: ChatGPT (OpenAI), Gemini (Google), Bard (Microsoft), Mistral, Llama, Falcon, Bloom, etc.

For the non-tech user, the easiest way to use AI was, until a few days ago, using a free version of a text LLM. That means, text in, text out. To get more, without going through the hassle of technical requirements, the solution was basically to pay to use a better model (ex: GPT4)

GPT-4o and multimodality

GPT-4o is the brand new openAI model. It was released a few days ago and will offer multimodality for free for all users.

What does it mean in our skills-building journey? Using an AI assistant and asking the right questions, you can get a lot more information than just reading your book. And way faster.

But above all, texts are not the only medium you’ll be able to use to sharpen your skills. Images, pdfs, videos, sounds… All of these can be handled, as input and output, by GPT-4o.

That means we could learn and interact with the AI over ebooks, graphs, podcasts, excel files, programming codebase… Let’s try that!

Real-life learning strategy, 2024 edition

Picture this: Iwant to be better at investing.

Now the first part of my old way of learning would be to make a list of books in my field of interest. Which would require me to browse Amazon’s (or any other marketplace for that concern) subcategories, on a topic I know almost nothing about.

How would I do with AI? First;

A simple prompt

I want to learn about investing. My current level is absolute beginner, but I’m into deliberate practice. As a beginner, what are the 5 first things I should learn about investing?

The answer gave me directions, in a few seconds:

1 — Understanding Financial Markets:

  • Key Concepts: Learn what financial markets are, how they function, and their types (stock markets, bond markets, commodities, and real estate).
  • Market Participants: Familiarize yourself with the roles of investors, companies, brokers, and regulators.
  • Market Indices: Understand major indices like the S&P 500, Dow Jones, and NASDAQ, and what they represent.

2 — Investment Vehicles and Asset Classes (will skip the bullet points for clarity)

3 — Risk and Return

4 — Investment Strategies

5 — Personal Finance and Investment Planning

Next.

Pick some books

Amazon search results for “financial markets” book. Screenshot by author.

Well… I’m glad AI will be able to help me. With 25 books on the 1st page only, tons of different reviews, points of view, prices… I would not know where to start.

Let’s start with the first category, “Understanding financial markets”. Please give me 5 must-read books tailored to my profile. For each of these books, tell me, in 2 to 3 sentences, why they are important and how it will help me move towards my goal of mastering investing

5 seconds later, here is my tailored list:

  • “The Little Book That Still Beats the Market” by Joel Greenblatt
  • “A Random Walk Down Wall Street” by Burton G. Malkiel
  • “The Intelligent Investor” by Benjamin Graham
  • “One Up On Wall Street” by Peter Lynch
  • “Market Wizards” by Jack D. Schwager
ChatGPT-4o list the books I must read, and why — Screenshot by author

From there, I could do many things: ask for more books or a refined list, get a small overview of the life and career of an author, get the average price for each of them, their length…

But let’s continue with our old-fashioned action list, step 3 is to “study the excerpts to set a logical and pedagogical order of reading them”.

I’m looking for a logical order of learning things. I can’t jump before I know how to walk. In what order would you advise to read the above books?

  • “The Little Book That Still Beats the Market” by Joel Greenblatt
  • “A Random Walk Down Wall Street” by Burton G. Malkiel
  • “The Intelligent Investor” by Benjamin Graham
  • “One Up On Wall Street” by Peter Lynch
  • “Market Wizards” by Jack D. Schwager

As you can see, the list is unchanged. Yes, the model understood, at the beginning, our goals and needs, and made sure its choices were justified.

The model explaining the order chosen — Screenshot by the author

From there, you can dive deep. You set the roadmap for your next days of learning, the AI knows your goals and will adapt to your progress. Here are a few other ideas you can try:

Please tell me more about Benjamin Graham. Why is he important in the economical landscape, and what are his main ideas?

Besides Warren Buffet, what are some successful investors in the financial markets and what can I learn from their strategies?

What is the “PER”? What does it stand for and how is it calculated? What roles does it have in stock picking strategy?

Using images

Text is not everything. Let’s try to upload an image I could have seen inside a book, and ask for more precision about it.

A chart from Ben Graham’s “The Intelligent Investor” — All rights reserved

I’m currently reading the third book, “The Intelligent Investor”. I don’t understand this chart. Can you help me, pedagogically? Please note that I only read the first 2 books. Then, sum up the goal of this chart in 2 to 3 sentences.

The model will guide you step by step through all items of the chart, and conclude along these lines:

The goal of the graphic is to illustrate the long-term performance of Benjamin Graham’s “Defensive Investor” strategy compared to the S&P 500 index from 1962 to 2012. It demonstrates that the defensive strategy, based on conservative investment criteria, yields significantly higher annualized returns (14.5%) than the S&P 500 (10%), highlighting the effectiveness of disciplined, value-based investing.

One step further

You now grasp the potential of using GPT-4o as a sidekick. Ready to go deeper with more scenarios?

I won’t specifically link them to the “1-hour rule” theory from now on but rather guide you through how you could integrate multimodality into your learning journey.

Images

  • Analyze the image of a non-Latin language (Japanese, Chinese, Korean, etc.) text. Colorize the adjectives and verbs, then render a new image. Convert the image to text, and break it down to learn the grammar and structuration
  • Upload the last photograph you took while learning about composition. Ask for feedback about framing and lighting. Improve your skills by applying the advice and settings.

Audio

  • Upload the file of a podcast you’re interested in. Automatically transcribe it, then sum it up. Extract the 5 key points, and ask why they are important. Explore if other prominent speakers of the field have similar or opposite ideas, then study them.
  • Upload the audio of a song, and ask the model to generate a partition for the instrument you’re learning. Ask for similar songs you can learn next once you master this one.
  • Converse with the IA directly in the language you’re learning and ask for correction. The model can also adjust its level to yours, and follow your progression

Video

  • Practice speaking facing the camera for your next presentation. Have the model analyze your posture, speed, expressions, and intonations. Ask for general advice, then dig deeper asking for links to renowned speakers’ techniques about what can be improved. Rinse and repeat.

Final words

These are only a few examples of what can be done. Remember that you’re the one shaping your path to success. Learning is a (long) journey, but now it can be efficient, fun, and rewarding.

As with many things AI-related, don’t take everything for granted. Ask for confirmation, read multiple sources, learn from professionals, take lessons, get mentored…

AI is not and will not be replacing human interaction, so keep interacting with others. Happy learning, on the road to mastering your future skill!

Tags:

No responses yet

Leave a Reply

Your email address will not be published. Required fields are marked *