Google DeepMind Loses Two Top Researchers in One Week
0xBroomberg
Google DeepMind lost Transformer co-author Noam Shazeer to OpenAI and Nobel chemistry laureate John Jumper to Anthropic within one week. This means → the AI talent war has escalated from poaching to direct lab-to-lab defections, putting systematic pressure on Google's ability to retain its best people.
Who left, and why do these two matter so much?
Noam Shazeer co-authored the 2017 paper *Attention Is All You Need*, which introduced the Transformer architecture — the "T" behind ChatGPT. In plain terms = without that paper, today's large-model wave does not exist.
Google previously spent over $2 billion acquiring Shazeer and part of his Character.ai team. That massive retention bet failed to create a lasting lock-in. This means → money alone is not enough; researchers' decisions hinge on where they see the future heading.
John Jumper won the 2024 Nobel Prize in Chemistry for AlphaFold, making him a flagship figure for DeepMind's scientific reputation. His move to Anthropic is a brand blow even more than a technical loss.
Why did they leave, and where did they go?
Shazeer joined OpenAI; Jumper joined Anthropic. Both are direct lab-to-lab moves — no startup detour, no acqui-hire. This reflects rivals surgically targeting Google's core talent, one by one.
Per Axios, top researchers weigh multiple factors: potential financial upside, access to compute, each company's competitive outlook, and whether leadership will deploy the technology responsibly.
On the financial side, Anthropic and OpenAI both face upcoming IPOs whose potential upside may exceed what already-public Google can offer. In plain terms = pre-IPO equity options can outweigh Google's cash packages.
What other talent moves happened at the same time?
Nvidia acquired the Essential AI team, bringing in Ashish Vaswani — another co-author of *Attention Is All You Need*. This means → the original Transformer author group is rapidly scattering across rival camps.
Barret Zoph returned to OpenAI after leaving Thinking Machines over "alleged misconduct," then departed again; his next destination is undisclosed. The talent churn cycle now runs on a weekly clock.
What makes elite researchers so valuable?
Per Axios, the most sought-after researchers offer far more than papers or code. They bring the intuition to judge which directions are worth pursuing, the experience to run large-scale experiments, and the gravitational pull to attract other top scientists.
In plain terms = one key hire can trigger a chain of follow-on recruits; one major departure can trigger a chain of exits. The talent war is fundamentally about capturing the person others follow.
Why is the competition especially fierce right now?
The industry broadly believes the window for AI models to achieve "recursive self-improvement" — models autonomously enhancing their own capabilities — is approaching. This means → the current period is seen as the decisive stretch; whichever lab holds the strongest researchers is most likely to dominate this inflection point.
A researcher's choice of employer is, at its core, a bet on which organization will lead the AGI moment. Whether Google can hold its position in the race under sustained talent attrition is the key variable to watch over the coming quarters.
Content is for reference only, not financial advice.