Aidan Gomez helped write the blueprint for modern generative AI. As a co-author of the 2017 paper Attention Is All You Need, he was part of the Google Brain research team that introduced the Transformer architecture. That technology now underpins today’s large language models.
The significance of the work became clear only later. Transformers allowed researchers to train powerful language models at far greater scale. They changed how machines process and generate language. The architecture now sits behind many of the systems driving the generative AI boom, including GPT and other large language models.
In 2019, Aidan Gomez co-founded Cohere with a practical question: what happens when these models leave the laboratory and enter the enterprise? As CEO, he has built the company around AI for business, with a particular focus on security, privacy and control.
That focus has become increasingly important as companies move AI from experimentation into core operations. Cohere develops language models, retrieval systems and AI agents that work within existing enterprise infrastructure. Its North platform brings these capabilities together to automate knowledge work. Its models can also run in private, dedicated and air-gapped environments.
Gomez has also broadened the conversation beyond model performance. His work increasingly focuses on the infrastructure and governance that surround AI. He has become an advocate for sovereign AI: giving governments and enterprises access to advanced models while retaining control of their data, infrastructure and deployment.
That thinking now shapes Cohere’s expansion. In 2026, the company announced plans to combine with German AI firm Aleph Alpha. The proposed combination would create a transatlantic AI business focused heavily on governments and regulated industries. It reflects a broader shift in the AI race. Model capability still matters, but so do security, jurisdiction and institutional control.
Before Cohere, Aidan Gomez worked at Google Brain alongside Jeff Dean, Jakob Uszkoreit and Geoffrey Hinton. He later pursued doctoral research at Oxford under Yarin Gal and Yee Whye Teh. He also led researchers at the independent AI group For.ai.
