Google researchers developed the Transformer architecture used by modern large language models.
The Transformer was introduced in the 2017 research paper “Attention Is All You Need,” written by researchers at Google Brain and the University of Toronto. Its central innovation was an attention mechanism that allowed a model to weigh relationships between words or tokens efficiently, without processing a sentence strictly one step at a time.
Transformers became the foundation for many influential language and generative-AI systems. They are used in models for text, images, audio, and multimodal tasks, although later systems often add substantial engineering beyond the original architecture.
A common mix-up is crediting OpenAI, because OpenAI popularized Transformer-based systems such as GPT. OpenAI built important models, but the original Transformer architecture was introduced by Google-led researchers.