OpenAI Hires Google DeepMind's Noam Shazeer and Former White House AI Official Ahead of IPO

OpenAI is adding two prominent figures to its ranks ahead of a planned public offering: Noam Shazeer, a longtime Google researcher and co-lead on the Gemini project, and Dean Ball, a former Trump administration AI policy official.

Shazeer announced his departure from Google on Wednesday after a tenure stretching back to 2000, interrupted only by a three-year period during which he co-founded the AI role-playing startup Character AI. Google rehired Shazeer two years ago through a $2.7 billion deal that also gave the company access to Character AI's technology.

His credentials in the field run deep. Shazeer is one of the co-authors of the seminal 2017 paper "Attention Is All You Need," which introduced the Transformer architecture that underpins modern generative AI systems.

Before his departure, Shazeer had drawn internal scrutiny at Google for comments he made on the company's messaging boards regarding transgender identity and Israel's war in Gaza, according to The Information. Management deleted those posts.

Ball, meanwhile, is set to join OpenAI on July 6 as the head of a newly created team called Strategic Futures, reporting directly to Chief Strategy Officer Jason Kwon.

"I am pleased and honored to announce that, on July 6, I'll be joining OpenAI as leader of a new team called Strategic Futures," Ball wrote on X. "Our mandate will be to help the company's leadership shape frontier AI policy."

In a blog post, Ball described the team as "small, high-agency" and said it would focus on catastrophic risk, recursive self-improvement, labor market impact, and the relationship between frontier AI labs, governments — particularly the U.S. federal government — and society. The team will handle both public-facing policy work and internal governance.

Ball had a brief stint in the White House last year, where he helped publish the administration's AI Action Plan before departing to return to the techno-libertarian think tank the Foundation for American Innovation as a senior fellow.

"Almost by necessity," Ball wrote, AI labs will have to take the lead on AI governance decisions, adding that "internal governance will be more central to the future of AI than most people realize."

The dual hires come as OpenAI's chief competitor Anthropic faces significant regulatory headwinds. Late last week, President Donald Trump ordered an export control ban on Anthropic's latest models, Fable 5 and Mythos 5, forcing the company to suspend access to both to avoid noncompliance.

The moves reflect a broader pattern of talent reshuffling across the leading AI laboratories, including Google, OpenAI, Anthropic, and Meta, as each organization positions itself ahead of what is expected to be an intensifying competition for regulatory favor in Washington.

OpenAI's decision to stand up a dedicated Strategic Futures team — and to staff it with a figure who until recently held a seat inside the White House — signals that the company views its relationship with the federal government as a strategic asset in its own right, not merely a compliance obligation.

How Shazeer will be deployed within OpenAI's research or product organization has not yet been disclosed.

Disclaimer