Alphabet loses $225 billion as AI talent exodus to OpenAI and Anthropic sparks investor fears

Alphabet Stock Faces Steepest Decline in Over a Year

Alphabet shares tumbled 7% on Monday, marking the company's worst single-day stock decline in more than a year and wiping out over $225 billion in market capitalization 1

1

3

3

. The stock plunge came as concerns about AI intensified following high-profile exits from Google DeepMind to rival companies. Alphabet stock underperformed both the Nasdaq and its megacap peers, with the decline far outpacing losses at Meta, which fell 2%, and Amazon, which dropped about 4% 4

4

. The selloff reflects growing investor concerns about Google's ability to retain top AI researchers amid the intensifying AI talent war.

High-Profile Exits Trigger Market Panic

The AI talent departures began when Noam Shazeer, Google's vice president of engineering and co-lead of its Gemini AI models, announced Wednesday that he was leaving to join rival OpenAI 1

1

. Shazeer's exit proved particularly costly given that Google had paid roughly $2.7 billion to bring him back from Character.AI less than two years earlier, in August 2024 5

5

. The researcher co-authored the pivotal 2017 paper "Attention Is All You Need," which introduced the Transformer architecture—the "T" in ChatGPT 2

2

. Just days later, John Jumper, who shared the 2024 Nobel Prize in Chemistry for AlphaFold, announced he was also leaving Google DeepMind for Anthropic after nearly nine years at the company 2

2

3

3

.

Race for AI Talent Intensifies Across Industry

The back-to-back departures highlight how the race for AI talent has intensified at the highest levels, with many AI developers believing that artificial general intelligence (AGI) is on the horizon 2

2

. According to Gil Lauria, head of tech research at DA Davidson, the high-profile exits "are raising the concern that Google is losing the war for talent at the frontier of AI" 4

4

. Companies are placing extraordinary value on a tiny number of humans who know how to direct AI research work, even as they race to automate more of that research itself 2

2

. The most coveted AI researchers offer more than technical knowledge—they bring judgment about which ideas to pursue, experience running enormous experiments, and the ability to recruit other sought-after scientists 2

2

.

Competitive Edge Concerns Mount for Google

Investors worry that startups such as Anthropic and OpenAI are offering AI researchers greater freedom, attractive compensation packages, and stronger incentives, making it harder for Google to retain top talent 3

3

. For top researchers, deciding where to work involves a complicated calculus: potential financial rewards, access to computing power, each company's prospects of leading the field, and whether its leadership will wield that power responsibly 2

2

. On the financial front, Anthropic and OpenAI have the advantage of forthcoming IPOs, which could offer greater upside than publicly traded companies like Alphabet 2

2

. Lauria noted that while Google had the state-of-the-art model for a few weeks last year, "it has fallen off since, and these departures may mean it is falling behind" 4

4

.

Heavy AI Spending Adds Pressure

Beyond the AI talent exodus, investors remain concerned about the pace of AI-related spending at tech giants. Google has signaled plans to spend $180 billion to $190 billion in fiscal 2026, with much of that money earmarked for AI compute and data centers 4

4

5

5

. Dave Wagner of Aptus Capital Advisors observed that "the market is drawing a sharp line between AI spenders and AI earners," noting that the big spending hurts hyperscalers' margins 4

4

. Alphabet is also facing an ongoing dilutive overhang from a record $84.75 billion equity raise, which has raised concerns about a pause in share buybacks 5

5

. Many in the industry see just a short window until AI models start getting better on their own through recursive self-improvement, making current talent retention decisions critical for long-term positioning 2

2

.