VC
Value Add VC
โšกHomePulseโšกHelpful Apps๐Ÿ“Blog๐ŸคPartner
Illustration for: Nobel Laureate John Jumper Leaves Google DeepMind for Anthropic, Deepening Google's AI Talent Drain
Value Add VC/Pulse/AINobel laureate hire

Nobel Laureate John Jumper Leaves Google DeepMind for Anthropic, Deepening Google's AI Talent Drain

John Jumper -- the 2024 Nobel laureate who co-created AlphaFold -- is leaving Google DeepMind after nearly nine years to join rival Anthropic. The exit lands just days after Gemini co-lead Noam Shazeer departed for OpenAI, turning a brutal week of poaching into the clearest sign yet that the AI race is now a talent war, and that Google's crown-jewel lab is the one bleeding.

By the Numbers

John Jumper
Departing
DeepMind โ†’ Anthropic
From โ†’ To
~9 years
Tenure
Chemistry, 2024 (AlphaFold)
Nobel
Shazeer โ†’ OpenAI
Prior Exit
TC
Trace Cohen
Early-stage VC & angel ยท Founder, New York Venture Partners
June 20, 2026
2 min read
ShareXLinkedInEmail

THE RUNDOWN

1

A sitting Nobel laureate defecting is a reputational blow money alone can't paper over -- it signals where top scientists think the frontier is moving

2

Back-to-back losses (Jumper to Anthropic, Shazeer to OpenAI) expose how fragile even DeepMind's roster is once momentum shifts

3

Anthropic, fresh off reportedly passing OpenAI on revenue, is now winning the people as well as the dollars

4

Talent concentration -- not just compute -- is becoming the decisive moat in frontier AI

TC

The VC Read ยท Trace's Take

Trace Cohen

Compute gets the headlines, but the real bottleneck in frontier AI is the few hundred people who can actually move a model -- and they vote with their feet. A Nobel laureate leaving DeepMind for Anthropic the same week Gemini's co-lead leaves for OpenAI isn't a coincidence; it's a momentum signal, and momentum is the whole game in recruiting. Anthropic passing OpenAI on revenue and now poaching the science talent is the flywheel founders should study: win the dollars, win the people, win the next dollars. The thing I'd watch is whether Google responds with structure and autonomy or just bigger checks -- because for this tier of person, the checks are already absurd and it clearly wasn't enough.

๐Ÿค– AI Landscape โ†’๐Ÿ“ˆ AI Valuations โ†’

Analysis

John Jumper, the Google DeepMind vice president who shared the 2024 Nobel Prize in Chemistry for the AlphaFold protein-folding breakthrough, announced on June 19 that he is leaving the company after nearly nine years to join Anthropic. It is one of the most senior and symbolically loaded defections of the entire AI cycle: not a junior researcher chasing a pay bump, but a Nobel laureate and the public face of DeepMind's most celebrated scientific achievement.

The timing makes it worse for Google. Jumper's move comes just days after Noam Shazeer, a co-lead on Gemini and one of the original 'Attention Is All You Need' authors, left for OpenAI. Two marquee names walking out of the same lab in one week reads less like normal churn and more like a confidence problem at the place that was supposed to be Google's answer to every rival.

โ€œJumper's move comes just days after Noam Shazeer, a co-lead on Gemini and one of the original 'Attention Is All You Need' authors, left for OpenAI.โ€

For Anthropic, the hire is a flex on top of a strong quarter. The company has reportedly seen its revenue run rate surge past OpenAI's on enterprise demand, and landing a Nobel laureate signals that the scientific frontier -- not just the commercial one -- is bending its way. Top researchers follow conviction and compute, and Anthropic is increasingly where both seem to be pooling.

The deeper story is that talent, not GPUs, may be the scarcest input in frontier AI. There are only a few hundred people on earth who can meaningfully push a frontier model, and the labs are now competing for them with nine-figure packages and equity in pre-IPO rockets. Every senior departure compounds: it takes knowledge, recruits more talent in its wake, and dents the morale of those who stay.

For founders and investors, the read-through is about durability. A lab's valuation rests on its people as much as its model weights, and this week showed how quickly that base can erode. Watch whether DeepMind can stem the bleeding -- and whether Anthropic's pre-IPO momentum, now in both revenue and talent, forces Google and OpenAI to respond before their own listings.

ShareXLinkedInEmail

More on

Anthropic โ†’OpenAI โ†’Google โ†’Google DeepMind โ†’

Analysis and editorial commentary by Value Add Pulse.

โ† Back to Pulse

THE WIRE in your inboxโ€” Tech, startup & VC news with Trace's take. Free, no spam.

Read Next

AIยท Aug 4, 2026

What OpenAI's Price Collapse Really Signals

Illustration for: What OpenAI's Price Collapse Really Signals
AI

What OpenAI's Price Collapse Really Signals

OpenAI cutting its cheapest model's price 80% three weeks after launch isn't just competitive pressure -- it's a company telling enterprise customers compute cost finally matters more than capability.

AIยท Aug 4, 2026

A Verified Math Proof and What It Means for AI Research Money

Illustration for: A Verified Math Proof and What It Means for AI Research Money
AI

A Verified Math Proof and What It Means for AI Research Money

OpenAI's Astra generated a machine-checkable proof of a decades-open math problem for roughly $2,000 in compute -- a result Fields Medalist Tim Gowers said he'd back for publication without hesitation, and a preview of what 'AI research spend' can now buy.

AIยท Aug 4, 2026

Humanoid Robots: Follow the Manufacturing Data, Not the Demos

Illustration for: Humanoid Robots: Follow the Manufacturing Data, Not the Demos
AI

Humanoid Robots: Follow the Manufacturing Data, Not the Demos

Humanoid robot startups have raised $8.6B in 2026 alone, already 1.8x all of 2025 -- but the more useful signal is production data: Figure is shipping past 1,000 units and AgiBot has 15,000 cumulative units in the field.

@Trace_Cohenยทt@nyvp.com