Just days after Google's vice president of engineering and Gemini AI co-lead, Noam Shazeer, announced his departure for OpenAI, DeepMind's vice president John Jumper also left for Anthropic. These rapid, high-profile exits triggered an immediate 5% drop in Alphabet's stock on June 22, 2026, according to Eciks. The departures signal a significant shift within the competitive landscape of artificial intelligence.

Google has historically been a crucible for groundbreaking AI research, fostering the very talent that built its advanced models. Yet, it is now losing the architects of its most sophisticated systems to the well-funded startups it helped inspire. This reverses the expected flow of talent in the industry, raising questions about Google's long-term AI leadership.

Based on this exodus of key AI researchers, Google will likely face increasing challenges in maintaining its leadership in frontier AI development. This trend could potentially cede ground to more agile and talent-rich competitors like OpenAI and Anthropic.

Who's Leaving and Where They're Going

  • AI researcher Noam Shazeer announced he was leaving Google for OpenAI, according to TechCrunch.
  • AI researchers Jonas Adler and Alexander Pritzel are leaving Google for Anthropic, as reported by TechCrunch. Separately, Bloomberg states these same researchers are planning to leave for Anthropic, indicating their departure status might still be in flux.

These confirmed and planned departures highlight a clear pattern of Google's top AI talent migrating to its direct competitors. This movement strengthens the research capabilities of rival firms, suggesting a systemic issue in Google's retention strategy rather than just isolated incidents.

The Architects Behind Google's AI

Noam Shazeer, Google's vice president of engineering and a co-lead for the Gemini AI project, announced his move to OpenAI on June 18, according to Eciks. Just one day later, on June 19, John Jumper, DeepMind's vice president and engineering fellow, also announced his departure after nine years to join Anthropic, as reported by Eciks. These individuals held pivotal roles in developing Google's most advanced AI models.