---
title: "ChatGPT: what it is, use cases and limitations"
pubDate: 2022-12-02
description: "In today’s blog post we will introduce you to ChatGPT - the latest project published by OpenAI. We will explain what it is, suggest some use cases but also mention important limitations."
author:
    name: 'Patryk Lasek'
    image: '/blog-assets/authors/patryk_lasek.jpeg'
image:
    url: '/blog-assets/posts/gptUseCases_bg.png'
    alt: "Background image for ChatGPT: what it is, use cases and limitations"
tags: ['insights']
---

## What is ChatGPT?

[ChatGPT](https://openai.com/blog/chatgpt) is the latest innovation from
[OpenAI](https://openai.com/). It is a model trained using Reinforcement
Learning from Human Feedback to interact in a **conversational way**. It is
fine-tuned from a model in the GPT-3.5 series, which finished training in
early 2022.

‍

## How does it work?

ChatGPT is designed to provide users with an intuitive and natural
conversation partner. It uses a range of features, including natural language
understanding, context-aware response generation, and natural language
generation. This makes it [capable](https://quickchat.ai/post/chat-gpt-in-businesses) of:

  * responding to questions,
  * carrying out tasks,
  * providing helpful guidance.

‍

## Use cases

ChatGPT is designed to be used in a variety of ways. It can be used as a
research consultant, brainstorming partner or a writing assistant. It can also
be helpful in providing personalized guidance during online shopping or even
for entertainment.

By far the best way to really understand how it works is to [**try it
out**](https://chat.openai.com/). If you need some more inspiration, check out
these mind-blowing examples shared on Twitter:

  * [ChatGPT writes an academic essay](https://twitter.com/corry_wang/status/1598176074604507136)
  * [Another essay on the history of the music scene](https://twitter.com/stuartbduncan/status/1598457158949130241)
  * [ChatGPT explains a bug, fixes it and explains the fix](https://twitter.com/amasad/status/1598042665375105024)

‍

## Limitations

The OpenAI announcement [blog post](https://openai.com/blog/chatgpt) lists
some limitations of the model including:

  * plausible-sounding but incorrect answers,
  * overusing certain phrases,
  * being sensitive to tweaks to the input phrasing.

Another interesting point was raised in a
[tweet](https://twitter.com/paulg/status/1598306206661107713) by Paul Graham,
the founder of Y Combinator. Given enough examples, large language models can
be really good at **imitating experts** or producing solutions that seem
**very convincing**. However, what is the path that takes us from impressive
demos generating really good drafts to actual business products that can be
truly relied upon? We explored this subject in our [previous blog
post](https://quickchat.ai/post/3-hard-truths-about-generative-ai).

‍

## How can I start using ChatGPT in my business?

ChatGPT will soon likely be available via API like other OpenAI models. **How
can you implement a large language model in your business?** That’s what we’ve
become experts in at [Quickchat AI](https://quickchat.ai/) over the past few
years.

The Quickchat AI platform allows you to build your custom conversational AI
experience on top of the best large language models like GPT-3. by customizing
your AI’s:

  1. **Personality** (the _style_ of conversation)
  2. **Knowledge Base** (what your AI _knows_ during the conversation)
  3. **Integrations** (interact with any internal or external system from within the conversation)

Our customers (including Robotics companies, law firms, startups and large
organisations) implement Quickchat AI as a **conversational interface** to
their products or as an **AI Expert system** to operate internally or
externally. If you would like to discuss how our technology could be
implemented for your use case, [please get in
touch](https://quickchat.ai/contact).
‍