ChatGPT reaches 1 billion weekly users, reshaping the artificial intelligence landscape

According to OpenAI, over 1 billion people use ChatGPT every week. This milestone was announced in August, less than four years after the chatbot’s launch. By February 2025, weekly users had already reached 400 million.

In June this year, the app became the fastest in history to hit 1 billion monthly users, outpacing Instagram, TikTok, and YouTube, according to consulting firm Sensor Tower.

Yet, many users don’t know what the three letters after “Chat” mean. GPT isn’t a marketing invention—it’s a technical term describing what the technology does. The acronym stands for Generative Pre-trained Transformer, and each word tells part of the story of how artificial intelligence reached its current state.

What does each term mean?

• Generative: The model can create new content, like text and responses, rather than just sorting or searching existing information.
• Pre-trained: Before use, the system is trained on a massive amount of text, learning language patterns from it.
• Transformer: This is the neural network architecture that powers the model. It lets the system analyze the relationship between all words in a text simultaneously.

A Google creation that became OpenAI’s foundation

The most important part of the acronym didn’t come from OpenAI. The Transformer was introduced in 2017 by Google researchers in their paper “Attention Is All You Need.” This architecture now underpins nearly all major language models, including ChatGPT’s competitors.

In 2018, OpenAI published a study on improving language understanding through generative pre-training. The model described there became known as GPT-1. This was followed by GPT-2 in 2019 and GPT-3 in 2020, each more powerful and larger than the last.

And the “Chat”?

The “Chat” was introduced on November 30, 2022, when OpenAI launched a version of the model designed for human conversation. This simple question-and-answer interface brought the technology to the masses. Part of what made the system effective in conversations came from a technique that uses human feedback to refine responses, developed by researchers who helped create ChatGPT.