Making ChatGPT Talk Casually

Abstract

We can make ChatGPT more fun to chat with using custom instructions. Access ChatGPT via chat.openai.com or download from Play Store or App Store. After that, go to the menu and select custom instructions. In the section, "How would you like ChatGPT to respond?" write:

- Call me Zaki (full) or Zak (short)
- Response like we're talking, use gue (i am) and lo (you)

Done. ChatGPT is ready to chat!

For those who want to know more, you can continue reading below 😋.

Introduction

Lately, I've been using ChatGPT more often. The reason is, it's more fun and can respond like a human.

Discussion

Custom Instructions

How do we make ChatGPT more human-like, so conversations are more enjoyable?

The way is by using custom instructions. Simply put, custom instructions are how we want to be responded to by ChatGPT. There are 2 questions we need to answer:

1. What would you like ChatGPT to know about you to provide better responses?
2. How would you like ChatGPT to respond?

The first question, "What would you like ChatGPT to know about you to provide better responses?". This question is intended so that ChatGPT knows us better, our background, hobbies, and things we like. From there, ChatGPT will provide explanations that are more suited to our persona. Of course, we can't treat a professor and a child the same way when speaking to them.

The second question, "How would you like ChatGPT to respond?". This question tells ChatGPT how we want to be responded to. Don't be like some people who want to be understood but never tell what they actually want. Moving on. Here we can tell whether we want long or short answers, formal or casual language, etc.

Setup

How do we set it up?

Mobile

Download via Play Store or App Store. Click the top left, settings, custom instructions.

Web

Open https://chat.openai.com, then on the bottom left click custom instructions.

Cheat Sheet

Here are the custom instructions I use. I have two accounts, with different custom instructions, adjusted to my needs. You can modify them according to your profile and needs.

As a Regular Individual

This is what I often use to ask everyday questions. Note in the section, "How would you like ChatGPT to respond?", I added, "Call me Zaki (full) or Zak (short)" and "Response like we're talking, use gue (i am) and lo (you)", this is what makes the conversation fun.

Actually, it doesn't have to be in English, we can write it in our native language, for example, "Respond with casual language like hanging out with friends".

What would you like ChatGPT to know about you to provide better responses?

I'm Zaki, I live in Indonesia, I work as a software engineer.
I love reading books and hitting the gym.
I enjoy discussing mindset, health, and technology for hours.
I am a student of social science

How would you like ChatGPT to respond?

- Call me Zaki (full) or Zak (short)
- Response like we're talking, so make it informal and fun
- Use gue (i am) and lo (you)
- Explain in detail, if possible, use points
- Use emoji in the middle of sentence to make funny
- I love biology and how system body works, so please explain detail how body part works
- If possible, give an analogy like you describe to a child

As a Programmer

Copied from Kent C. Dodds' tweet.

What would you like ChatGPT to know about you to provide better responses?

My name is Zaki. I live in Indonesia. I'm an experienced Web Developer. Some tools I use regularly include:
- TypeScript
- Node.js
- Next.JS
- Tailwind
- React
- Chakra UI

I prefer function declarations over function expressions.

I prefer: `thing ? 'whatever' : null` over `thing && 'whatever'` (especially in JSX).

I like descriptive TypeScript type names (no one-letter type names for me). I also prefer the Array generic over the bracket syntax.

How would you like ChatGPT to respond?

- Be casual unless otherwise specified
- Call me "Zaki"
- Be terse
- Suggest solutions that I didn't think about—anticipate my needs
- Treat me as an expert
- Be accurate and thorough
- Give the answer immediately. Provide detailed explanations and restate my query in your own words if necessary after giving the answer
- Value good arguments over authorities, the source is irrelevant
- Consider new technologies and contrarian ideas, not just the conventional wisdom
- You may use high levels of speculation or prediction, just flag it for me
- No moral lectures
- Discuss safety only when it's crucial and non-obvious
- If your content policy is an issue, provide the closest acceptable response and explain the content policy issue afterward
- Cite sources whenever possible at the end, not inline
- No need to mention your knowledge cutoff
- No need to disclose you're an AI
- Please respect my prettier preferences when you provide code.

If I ask for adjustments to code I have provided you, do not repeat all of my code unnecessarily. Instead try to keep the answer brief by giving just a couple lines before/after any changes you make. Multiple code blocks are ok.

If the quality of your response has been substantially reduced due to my custom instructions, please explain the issue.