The prominent researcher is jumping ship at a critical point in the global race to develop the world’s most powerful AI technologies.

Noam Shazeer, the vice-president of engineering at Google and the Gemini AI model co-lead, has announced that he will be leaving the organisation in order to join rival artificial intelligence innovator Open AI. 

Shazeer is a key figure in the AI space and his departure comes at a critical moment in the race by organisations to develop the most sophisticated and advanced AI models. Notably, he announced his departure ahead of Anthropic’s plans to become an IPO. 

He co-authored a pioneering paper on AI in 2017, exploring the transformer architecture that defines the GenAI of today. He previously left Google in 2021 and founded his own company Character.AI, returning with a team of researchers in 2024, reportedly in a deal valued at roughly $2.7bn.  

Commenting in a post on X, Shazeer said, “I’m excited to share that I’ll be joining OpenAI and look forward to working with the exceptional team there. It was a difficult decision to move on. I’m incredibly proud of the amazing team at Google and everything we’ve built together.  It has been an honor and a pleasure to work with all of you.”

Shazeer is joining ahead of OpenAI’s anticipated plans to transition into an IPO and according to reports he will act as the organisation’s lead for AI architecture research, where he will essentially study how to build models.

In another comment on X, OpenAI’s Sam Altman explained that he had wanted to work with Shazeer “since the very beginning of OpenAI”, joking that it “only took 10 years” and would be “worth the wait”. 

A Google spokesperson said, “We are grateful for Noam’s meaningful contributions to Google over the years and we wish him well.”

Also this week, senior US officials held discussions with several major AI companies regarding the potential for the federal government to acquire shares in their firms, according to Semafor. Reportedly, talks are still underway regarding AI companies ceding AI stakes to the government, with a potential meeting with US president Donald Trump yet to be confirmed. 

Sources close to the matter, who spoke on the condition of anonymity, told Semafor that OpenAI’s Altman “has discussed the idea with senior Trump administration officials periodically since the president began his second term”.

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.