It took just a few days for one of the most valuable companies on the planet to suddenly look vulnerable.
Alphabet, the parent of Google, shed roughly $225 billion to $270 billion (estimates vary by market data providers) in market value as its shares tumbled on Monday, June 22, 2026. It was the stock's worst single-day drop in more than a year. The slide capped weeks of volatility since the company touched a record high in mid-May. It was set off by a story that has less to do with quarterly earnings and more to do with people walking out the door.
In the span of barely 48 hours, Google lost two of the most celebrated names in artificial intelligence. Both were reportedly moving to rival AI labs.
Noam Shazeer, a vice president of engineering and co-lead of Google's Gemini models, announced he was leaving for OpenAI. His exit carried symbolic weight far beyond his title. Shazeer is one of the eight authors of the 2017 research paper that introduced the Transformer architecture, which underpins essentially every major modern AI system, including Gemini. He had only recently returned to Google, which acquired his startup Character.AI-related assets in a deal valued at around $2.7 billion. Less than two years later, he was gone again.
A day later, John Jumper was reported to be leaving Google DeepMind for Anthropic. Jumper is a leading researcher behind AlphaFold, a system that solved a decades-old problem in biology by predicting protein structures. Reports also suggested additional DeepMind researchers were exploring moves to other AI labs. The scale of departures remains limited relative to DeepMind’s overall research team.
For investors, the symbolism was difficult to ignore. Key contributors to some of Google’s most important AI breakthroughs were increasingly being recruited by competitors.
On their own, a handful of high-profile departures might have been a one-day headline. Instead, they landed on an already jittery market.
Google has been spending heavily to stay competitive in AI, with capital expenditures running into tens of billions per quarter. Full-year spending is expected to remain elevated. This has been funded in part through strong cash generation and external financing activity, but it has also raised concerns among some analysts about pressure on free cash flow as AI infrastructure costs scale.
Investor anxiety has also been shaped by broader industry debate. Some executives and analysts have warned that advanced AI models may become commoditized over time. Competition could shift toward pricing and distribution rather than pure capability. This has raised concerns that heavy investment may not translate into durable pricing power.
For investors, the combination of high spending and rising competition made the talent departures feel more significant than they might otherwise have been.
Not everyone is convinced the selloff reflects a structural problem.
Google's leadership pushed back on concerns about talent retention. They argue that movement between leading AI labs is a normal feature of a highly competitive research market. DeepMind leadership has similarly emphasized that the company continues to employ one of the largest AI research teams in the industry.
A number of analysts also viewed the selloff as overdone. Their argument rests on Google’s scale advantage: a global search and advertising business, integration across Android and YouTube, and a growing cloud division. Combined with custom AI chips (TPUs), this infrastructure may allow Google to deploy AI at lower cost and at massive scale.
Under this view, Gemini does not necessarily need to be the absolute best model in the world to succeed commercially. It only needs to be good enough to be distributed across Google’s ecosystem.
The week highlighted a broader question that continues to shadow the AI industry: is this a routine reshuffling of talent in a rapidly growing field, or a signal that leadership in foundational AI research is becoming more distributed?
Google remains one of the key origins of modern AI breakthroughs. The Transformer architecture and AlphaFold both emerged from its research ecosystem. An increasing number of researchers behind such advances have moved on to competitors, raising questions about how research leadership translates into long-term product dominance.
Comments