AI-Assisted Programming

Chapter 6. ChatGPT and Other General-Purpose LLMs

In this chapter, we’re going to dive into some widely used general-purpose large language models that can whip up code. Sure, they might not have all the bells and whistles like those that specialize in coding—such as with integration with IDEs, sophisticated security, or guardrails for the types of code used to train the models—but they’re still impressive. Plus, they’re not just about coding; they’re also great for planning and brainstorming.

We’ll check out the big-name chatbots like ChatGPT, Gemini, and Claude. I’ll walk through how to set them up, what they cost, what their cool features are, and how to get the most out of them.

ChatGPT

ChatGPT, short for Chat Generative Pretrained Transformer, exploded on the scene on November 30, 2022. OpenAI, the brains behind the app, didn’t really splash out on marketing. Turns out, they didn’t have to. ChatGPT quickly blew up and went viral all on its own.

In just five days, this app pulled in a million users, and within a couple of months, the number of users skyrocketed to 100 million. It became the quickest-growing platform in internet history.

Sandhini Agarwal, who helps to develop policy at OpenAI, noted, “I think it was definitely a surprise for all of us how much people began using it. We work on these models so much, we forget how surprising they can be for the outside world sometimes.”

The buzz kept growing. By September, ChatGPT had over 1.5 billion visits, and revenues had ...