ChatGPT Use Cases ~2 Months In

It has been just over two months since the release of ChatGPT. During this time, I have explored various use cases and compared different tools. In this post, I would like to share the two uses for ChatGPT that have proven to be the most valuable so far. Of course, it can be used for a thousand other things, but these two have personally stood out to me as the most consistently useful.

Use Cases

I have found that ChatGPT is most consistently useful in the role of an expert, tutor, and editor.

1. Expert and Tutor

When: I want to learn X.

Expertise vs Tutoring

These are very similar conceptually, but I would make a small distinction:

  • Expertise. At times, I may want to get a brief overview (tldr) of a topic. This could be viewed as a quick chat with an expert. For example, if I wanted a brief summary of the life extension industry, but didn’t want to dive too deep, I could think of it as a quick chat with a friend who works in the industry.
  • Tutoring: In other instances, I may want to dive deeper into a specific topic and ask a series of questions. For instance, while I was learning about machine learning with PyTorch, I found ChatGPT to be very helpful as a tutor.”

How

  • Know Where to Start. If I don’t know where to begin, a general introductory question might be helpful, such as “What are the most important topics to master in order to learn X?”
  • Troubleshoot. If I encounter a specific problem or question, it’s best to address it directly. I’ve found that this approach works particularly well with programming questions, at least at a basic level.

Best Practices

  • Creative Prompting: Sometimes, getting creative with prompts can help. For example, instead of asking, “Explain general relativity as if I’m five years old,” it may be helpful to try, “Explain general relativity as if I’m five years old. Please respond as a physics professor at MIT who is good at communicating complex ideas without dumbing them down.”
  • Follow-up: In my experience, one of the main advantages of ChatGPT over traditional search engines and previous-generation AI assistants or chatbots is its interactivity and ability to maintain the thread of the conversation. So, just like with a real tutor, it may be helpful to ask for explanations in a different way, provide an example, elaborate on a particular point, or step back and explain how a certain concept fits into the big picture.

3. Editor

When: I want to change anything about a text.

How

Some examples include:

  • Checking grammar
  • Summarizing or rephrasing the text
  • Getting ideas on what else to consider and think about
  • Analyzing the argument. For example, “What are the likely influences behind this text?” or “What are the strongest counter-arguments to this?”

Context

Lessons and Challenges

  1. Hallucinations and Validation. Of course, one must watch out for hallucinations. That’s why I find that using ChatGPT in combination with Google may work best. The first step may be asking ChatGPT to provide an initial overview of a topic to save time reading long articles from the Google’s organic results that have been SEO’d in and out. And the next step may be to search for specific sub-topics — sometimes using search syntax/operators to find exact matches or limit the search to certain websites.
  2. Common Knowledge vs. Advanced/Niche Topics: I’ve found that ChatGPT is very good at summarizing and explaining common non-controversial knowledge, but is more likely to hallucinate or struggle when dealing with advanced or niche topics that may only be covered by a just few papers or blog posts. Another way to think about this is to consider how likely it is that this topic was prominent in the training dataset.

What I’ve Tested

Despite still relying on Google for the majority of my searches, I have come to the conclusion that ChatGPT — especially when combined with Google — is better or comparable for a specific subset of my use cases and queries. Depending on the question or query, I have also found Perplexity.ai, Neeva, You.com/chat, and Elicit.org to be worth trying out.

What’s Next

I am looking forward to trying Microsoft’s integration of ChatGPT with Bing and Google’s integration of Bard with Google Search. I find the intersection of search and conversational AI to be particularly intriguing, as these two tools are some of the best resources for organizing and retrieving information that we have today.




Examples

Please see — or skip — the three examples of actual “conversations” below:

Continue reading

On Good Judgment and Decision-Making: The Science and Practice

Imagine you’ve made two decisions. One has resulted in a loss of 10 thousand dollars, while the other resulted in a gain of 10 thousand dollars. Would you say that the former was a bad decision, and the latter was a good one?

The first response that usually comes to everyone’s mind is “of course, the second decision was better”. But this is not necessarily the case. Can you think of a scenario where your response would be the opposite?

I’ve been thinking and reading about decision-making for many years while trying to put everything I learned into practice. Below, I summarize the strategies and mental models that I personally found most useful.

The Process vs. The Outcome

We all have a natural tendency to judge decisions based on their outcomes. This is not the worst heuristic, as there is a correlation between the quality of decisions and outcomes. But this heuristic has a major flaw — it doesn’t account for luck and incomplete information.

Continue reading

Learning How to Learn (And 20+ Studies)

I have been interested in cognitive science and effective learning methods for years. I’ve read multiple books and articles and put many ideas to test. In this post, I aim to synthesize my notes and provide references to scientific studies.

Summary (TL;DR)

Effective Learning Strategies

  • Distributed learning. Study less in each session but more frequently.
  • Active recall. Actively test your knowledge and skills.
  • Distributed recall. Space the tests in time and adjust the intervals based on performance.
  • Interleaving. Practice multiple related yet different skills/concepts simultaneously.
  • Elaborative interrogation (quiz-and-recall). Ask yourself questions and use the material you’ve learned to answer them.
  • Self-explanation and the Feynman technique. Explain what you’ve just learned in simple terms.

Physiology and Brain’s Health

  • Sleep
  • Exercise
  • Nutrition

Disclaimer and Introduction

I have no formal background in cognitive science or neuroscience, and this has been more of a side interest. My understanding is limited, and I still need to learn how to effectively and consistently apply all these ideas to practice.

That being said, I found some of the methods described in this article very useful. For example, I’ve used them to learn foreign languages, the basics of programming, and various disciplines covered during the two-year MBA program.

Effective Learning Strategies

Strategy #1: Distributed (Spaced) Learning Practice

In short, it’s better to distribute one’s practice over a period of time than cram it into one day.

In one study, elementary school students were asked to study in one of the three ways: massed, clumped, and spaced.

  • Massed = four lessons at a time
  • Clumped = two lessons on one day and two lessons on the next day
  • Spaced = one lesson per day for four days

The “spaced” group performed best, followed by the “clumped” group:

Another study compared comprehension scores under three different conditions:

  1. Read a text once (“single”)
  2. Read a text twice (“massed”)
  3. Read a text twice with a week-long gap (“distributed”)

When tested immediately, the second group performed best. But when tested with a delay of two days, the third group performed best.

This method is also superior for learning motor skills.

How to apply this in practice:

Create a learning schedule or find time to practice a little bit every day or every few days instead of cramming all your learning into one or just a few days.

If you’d like to learn more, read the Wikipedia article on distributed practice.

Strategy #2: Active Recall (Retrieval) Practice

It might be more effective to actively retrieve the information you’ve already learned than passively re-read or try to learn it once again.

One study that compared a method that emphasized study sessions with a method that emphasized tests and found the latter to be more effective for delayed recall.

  • SSSS = four study sessions
  • SSST = three study sessions, followed by one test
  • STTT = one study session, followed by three tests

Even imagining that you might be tested on the material you’re learning might help improve the recall.

How to apply this in practice:

If a few days ago you learned how past tense works in the Spanish language, try to remember the rules or even test yourself on your knowledge — instead of simply re-reading the same material once again.

You can read more about the active recall practice on Wikipedia.

Continue reading

Book Notes: How Google Works by Eric Schmidt

How Google Works” by Eric Schmidt, despite being quite self-congratulatory and maybe even prone to confirmation bias, is full of inspirational ideas and bids of practical wisdom to learn from. I took a few (ok, quite a few) notes on smart creatives, decision making, hiring, innovation, strategy, career, management and even managing email.

On smart creatives:

“And who, exactly, is this smart creative? A smart creative has deep technical knowledge in how to use the tools of her trade and plenty of hands-on experience. In our industry, that means she is most likely a computer scientist, or at least understands the tenets and structure of the systems behind the magic you see on your screens every day. But in other industries she may be a doctor, designer, scientist, filmmaker, engineer, chef, or mathematician. She is an expert in doing. She doesn’t just design concepts, she builds prototypes. She is analytically smart. She is comfortable with data and can use it to make decisions. She also understands its fallacies and is wary of endless analysis. Let data decide, she believes, but don’t let it take over.

She is business smart. She sees a direct line from technical expertise to product excellence to business success, and understands the value of all three. She is competitive smart. Her stock-in-trade starts with innovation, but it also includes a lot of work. She is driven to be great, and that doesn’t happen 9-to-5. She is user smart. No matter the industry, she understands her “get it right the next time around. She is self-directed creative. She doesn’t wait to be told what to do and sometimes ignores direction if she doesn’t agree with it. She takes action based on her own initiative, which is considerable.

She is open creative. She freely collaborates, and judges ideas and analyses on their merits and not their provenance. If she were into needlepoint, she would sew a pillow that said, “If I give you a penny, then you’re a penny richer and I’m a penny poorer, but if I give you an idea, then you will have a new idea but I’ll have it too.” Then she would figure out a way to make the pillow fly around the room and shoot lasers.

She is thorough creative. She is always on and can recite the details, not because she studies and memorizes, but because she knows them. They are her details. She is communicative creative. She is funny and expresses herself with flair and even charisma, either one-to-one or one-to-many.”

 

HIPPOs:

“Hippopotamuses are among the deadliest animals, faster than you think and capable of crushing (or biting in half) any enemy in their path. Hippos are dangerous in companies too, where they take the form of the Highest-Paid Person’s Opinion. When it comes to the quality of decision-making, pay level is intrinsically irrelevant and experience is valuable only if it is used to frame a winning argument. Unfortunately, in most companies experience is the winning argument. We call these places “tenurocracies,” because power derives from tenure, not merit. It reminds us of our favorite quote from Jim Barksdale, erstwhile CEO of Netscape: “If we have data, let’s look at data. If all we have are opinions, let’s go with mine.”

When you stop listening to the hippos, you start creating a meritocracy, which our colleague Shona Brown concisely describes as a place where “it is the quality of the idea that matters, not who suggests it.” Sounds easy, but of course it isn’t. Creating a meritocracy requires equal participation by both the hippo, who could rule the day by fiat, and the brave smart creative, who risks getting trampled as she stands up for quality and merit.”

 

On technical insights as a driver of innovation:

“Bet on technical insights, not market research. Product leaders create product plans, but those product plans often (usually!) lack the most important component: What is the technical insight upon which those new features, products, or platforms will be built? A technical insight is a new way of applying technology or design that either drives down the cost or increases the functions and usability of the product by a significant factor. The result is something that is better than the competition in a fundamental way. The improvement is often obvious; it doesn’t take a lot of marketing for customers to figure out that this product is different from everything else.

For example, at that time Google was experimenting in applying some of our expertise from online advertising to other advertising markets, including print, radio, and TV. These were clever efforts, supported by smart people, but they lacked that fundamental technical insight that would shift the cost-performance curve non-incrementally and provide significant differentiation. All three ultimately failed. And when we look back at other Google products that didn’t make it (iGoogle, Desktop, Notebook, Sidewiki, Knol, Health, even the popular Reader), they all either lacked underlying technical insights from the outset, or the insights upon which they were based became dated as the Internet evolved.”

Continue reading

Stanford Class: How to Start a Startup by YC

Free, open to everyone and highly educational Stanford class “How to Start a Startup” has just ended. But all the materials, including talks by star speakers, such as Paul Graham, Peter Thiel, Ben Horowitz, Sam Altman, Brian Chesky and others are going to be available online. For quick reference, here is the complete collection of all course materials:

Videos.

Lectures

Date Speaker Topic
9/23/14 Sam Altman, President, Y Combinator
Dustin Moskovitz, Cofounder, Facebook, Cofounder, Asana, Cofounder, Good Ventures
Welcome, and Ideas, Products, Teams and Execution Part I
Why to Start a Startup
9/25/14 Sam Altman, President, Y Combinator Ideas, Products, Teams and Execution Part II
9/30/14 Paul Graham, Founder, Y Combinator Before the Startup
10/2/14 Adora Cheung, Founder, Homejoy Building Product, Talking to Users, and Growing
10/7/14 Peter Thiel, Founder, Paypal, Founder, Palantir, and Founder, Founders Fund Competition is For Losers

Continue reading

Book Notes: What Technology Wants

Here are some notes from What Technology Wants by Kevin Kelly:

On inter-dependancies:

Each new invention requires the viability of previous inventions to keep going. There is no communication between machines without extruded copper nerves of electricity. There is no electricity without mining veins of coal or uranium, or damming rivers, or even mining precious metals to make solar panels.

On evolution of the scientific method:

The classic double-blind experiment, for instance, in which neither the subject nor the tester is aware of what treatment is being given, was not invented until the 1950s. The placebo was not used in practice until the 1930s. It is hard to imagine science today without these methods.

On games:

The cybernetician Heinz von Foerster called this approach the Ethical Imperative, and he put it this way: “Always act to increase the number of choices.” The way we can use technologies to increase choices for others is by encouraging science, innovation, education, literacies, and pluralism. In my own experience this principle has never failed: In any game, increase your options.

On sacrifices as a form of  investment:

As Suketu Mehta, author of Maximum City (about Mumbai), says, “Why would anyone leave a brick house in the village with its two mango trees and its view of small hills in the East to come here?” Then he answers: “So that someday the eldest son can buy two rooms in Mira Road, at the northern edges of the city. And the younger one can move beyond that, to New Jersey. Discomfort is an investment.

On using older technology as a statement:

Who would have guessed anyone would burn candles when lightbulbs are so cheap? But burning candles is now a mark of luxuriant uselessness. Some of our hardest-working technology today will achieve beautiful uselessness in the future. Perhaps a hundred years from now people will carry around “phones” simply because they like to carry things, even though they may be connected to the net by something they wear.

Continue reading

Book Notes: Surely You’re Joking, Mr. Feynman

Feynman is one of the brightest physicists of the 20th century who also happened to be a talented lecturer and an author of a book which is fun to read. Here are some excerpts from “Surely You’re Joking, Mr. Feynman” that I found interesting:

Surely, You're Joking, Mr. Feynman - book review & quotes

On freedom:

So I have just one wish for you–the good luck to be somewhere where you are free to maintain the kind of integrity I have described, and where you do not feel heed by a need to maintain your position In the organization, or financial support, or so on, to lose your integrity. May you have that freedom.

Isn’t it a modern luxury that very few of us can afford?

On pointless communication

When it came time to evaluate the conference at the end, the others told how much they got out of it, how successful it was, and so on. When they asked me, I said, “This conference was worse than a Rorschach test: There’s a meaningless inkblot, and the others ask you what you think you see, but when you tell them, they start arguing with you!

On mindfulness and watching your thoughts become dreams:

I also noticed that as you go to sleep the ideas continue, but they become less and less logically interconnected. You don’t notice that they’re not logically connected until you ask yourself, “What made me think of that?” and you try to work your way back, and often you can’t remember what the hell did make you think of that!

So you get every illusion of logical connection, but the actual fact is that the thoughts become more and more cockeyed until they’re completely disjointed, and beyond that, you fall asleep.

After four weeks of sleeping all the time, I wrote my theme, and explained the observations I had made. At the end of the theme I pointed out that all of these observations were made while I was watching myself fall asleep, and I don’t really know what it’s like to fall asleep when I’m not watching myself. I concluded the theme with a little verse I made up, which pointed out this problem of introspection:

Continue reading

Two Books About Brain And Neuroscience

Not a long time ago I wrote a post about My Stroke of Insight by Jill Bolte Taylor. And during last couple of months I listened to two more books: Brain Rules by John Medina and Your Brain at Work by David Rock. Unlike Jill, they don’t tell their own stories but try to give real life recommendations based on neuroscience research.

John focuses on general principles rules of brain functioning which he covers relatively briefly. David on the other hand provides more of a deep dive in various situations that we face daily, mostly at work but views them through the prism of our brain and its biochemistry. Social concepts, such as status, reward and others are explained through things like oxytocin, dopamine & epinephrine.

Those who find such topics interesting can find my notes below. Plus, a couple of great videos of authors’ talks and one fun Slideshare presentation.

1. Brain Rules by John Medina

I used the actual “brain rules” by John from his website as the basis for my notes and briefly tried to explain main idea of each one.

John Medina

EXERCISE | Rule #1: Exercise boosts brain power.

John recommends various kind of physical activity, especially aerobic one, including long walks. He states that if participants of business meetings walked on treadmills with 1.8 mile/hr speed, they would come up with much more creative ideas, not to mention increased memory and overall well-being. By the way, John takes his own medicine. It took him 15 minutes to adapt to replying to emails while walking.

Continue reading