A few years ago, I never imagined I’d spend part of my day prompting an AI model to help me write code, summarize meetings, or debug a flaky test. Back then, AI felt like something reserved for research labs. Sure, you’d hear about it beating humans at Go or generating surreal images, but it felt far from the day-to-day work of building software.
Then one day, it wasn’t.
When ChatGPT launched, for the first time, AI wasn’t hidden behind a PhD or a research API. Anyone could type into a box and get real, helpful answers.
That was the shift.
AI has been here all along
Let’s be clear: AI isn’t new. The term’s been around since the 1950s. Machine Learning has powered things like fraud detection, recommendations, and search results for years. But most of it lived behind the scenes. If you weren’t working at Google or Meta, you weren’t building AI, you were just consuming it.
Before ChatGPT, most AI has been narrow and specialized. You’d train a model to do one thing well: classify spam, detect faces, translate text. These systems were powerful, but they were expensive, required tons of data, and needed ML experts to build and maintain. The bar to entry was high.
So for most engineers? AI was something other people worked on.
The Rise of LLMs
In 2017, researchers came up with a new design for AI models called the Transformer. It made a big difference. Instead of just guessing the next word, it starts generating the next words with context awareness. That’s what made today’s AI feel a lot more helpful, more than just fancy autocomplete.
GPUs got more powerful and affordable, so training big models wasn’t just for huge tech companies anymore. At the same time, scraping large chunks of the internet became easier. That helped models like GPT could be trained on everything from books to Wikipedia to random blog posts, forum threads, and public code.
In 2020, OpenAI released GPT-3. It was impressive, but still mostly a toy for techies.
In late 2022, ChatGPT changed everything. Same core model, but better tuned and packaged with a simple chat interface. You didn’t need to understand all the complex Machine Learning concepts to use it. You just asked it a question.
The adoption number was wild. 100 million users in two months. It is faster than any app in history.
Suddenly, everyone, whether you were a student, a teacher, a marketer, a software engineer, or someone with no tech background at all, you could ask it a question or describe a task and see results. For the first time, people could personally experience what modern AI could do, without needing to know how it worked under the hood. This brought AI into the everyday lives of millions.