The Leverage of LLMs for Individuals

Disclaimer: This article is not meant to provoke anxiety or exaggerate the power of GPT. It is merely my personal observation after using ChatGPT/GPT-4 intensively for the past six months. It is definitely not applicable to the vast majority of people. This article is for those who wish to create something and have no expectations of “foreseeable returns” on an individual level.

GPT-4, not ChatGPT

First of all, GPT-4 and ChatGPT are two different entities.

If at this point in time (2023.05.10), there are still media outlets boasting or belittling GPT while using ChatGPT as an example instead of GPT-4, then it is not worth reading. The metrics provided by OpenAI’s official website are:

Source: https://openai.com/research/gpt-4

The recently released SuperCLUE Chinese large model benchmark shows that in terms of professional capability, GPT-4 is in a league of its own compared to other models, and this is just for Chinese.

https://github.com/CLUEbenchmark/SuperCLUE

Boundless Creativity

I had never written front-end code before, but after 48 hours of conversation with GPT-4, I built a podcast search website. I open-sourced the website and gave it a GPT-4.0 License - to express my gratitude.

PodFind - Acquire New Knowledge with Podcasts

A few days later, I wanted to automatically skip certain timestamps when watching videos on web pages. With no experience in Chrome extension development, I followed GPT-4’s instructions to create files, paste, and drag and drop, achieving it in less than 15 minutes. I didn’t publish it; it became a tool serving only me.

Skip the intense moments in horror movies with TimestampX

Whenever I have a new idea, I ask GPT-4 to write the most basic version. I provide feedback, it apologizes, and we gradually optimize to the 1.0 version I have in mind. I use up my GPT-4 quota (25 entries/3 hours) multiple times a day. With GPT-4’s support, I feel unstoppable. The overnight surge in productivity is intoxicating. It’s not about making money or starting a business, but purely about continuously bringing ideas from my mind into reality, which feels like happiness.

More importantly, it gives me the courage to dream and attempt things beyond my current abilities. I even plan to create a 3D game. I have a vague feeling: don’t limit yourself—soon, I might be using products developed by designers, lawyers, or electricians.

Recently, the two most frequent tasks I have used GPT for are:

  • Copying and pasting documentation and APIs to GPT-4, asking it to write interfaces based on them. Current documentation is written for human reading, with a mix of text and code blocks that makes copying a poor experience. Perhaps soon, documentation and APIs will become GPT-friendly first.

Having GPT write interfaces based on GPT official documentation

  • Translation and product localization. Even in the DeepL era, slightly longer sentences can still reveal machine translation traces. However, based on my experience, the translation capability of GPT (whether 3.5 or 4) far surpasses DeepL. I can localize my products into dozens of other languages at once.

Localizing into ten languages at once

Using GPT to create product preview pages in other languages

Within half a year, I launched 5 iOS apps.

Individual, not Company

Large language models are a leverage that can amplify the abilities of each user.

If you are an individual developer, the leverage it provides might be 10 times greater, but when you work for a company, that number might be only 2. So, even knowing the current economic downturn and wave of layoffs, for those who want to achieve ideas and create, I still want to express this bold opinion:

Staying in any company right now is a negative return; you are wasting personal leverage.

Unless you work for OpenAI, your GPT leverage is likely wasted on trivial business code, and even more likely that due to the overwhelming amount of poor code, GPT with its 4/8k context window is unable to optimize, further weakening the leverage.

I see too many talented young people trapped in large companies, exhausting their energy in endless documentation and meetings. Maintaining income is the main reason, but they may also be unaware of the potential they have when using GPT alone.

To be more extreme, I even think that joining any team is a kind of slowdown. Only by sailing alone in a small boat can one maintain agility.

Setting Sail Alone

How to generate ideas and sustain a livelihood is not the scope of this article. I still believe that this path is not suitable for the vast majority of people. What I’m doing is simply shouting towards those people who haven’t had the chance to use GPT-4 and haven’t considered the infinite possibilities it enables through its leverage.

I hope this is useful to the 0.01% of readers.