Transformer
The design, or architecture, behind almost every modern AI language model. Its great trick is paying attention to how all the words in a piece of text relate to one another, rather than reading strictly left to right. This ability to weigh context is called a transformer, and it is the breakthrough that made today's AI possible.
A transformer is the underlying structure that most current AI language models are built on. You do not need to know how it works to use AI well, but the name comes up so often that it helps to have a rough picture of what it is.
The core idea is something researchers call “attention.” When a transformer reads a sentence, it looks at every word in relation to every other word and works out which ones matter most for the meaning. In the sentence “the bank of the river,” it can tell that “river” is what fixes the meaning of “bank.” Earlier designs struggled with this kind of long-range connection, and the transformer handled it far better.
This design was introduced by researchers in 2017 and changed the field quickly. The “T” in GPT stands for transformer, and the same idea sits underneath Claude, Gemini, and most other well-known models. It is one of those rare technical ideas that genuinely marks a before and after.
For everyday use, the practical point is that this shared foundation is why so many different AI tools behave in broadly similar ways. They are, under the bonnet, variations on the same clever idea.