< Back to all clusters
[TECHNOLOGY] · United States, United Kingdom · 3 sources

started · updated

Google DeepMind faces decline in elite AI talent recruitment

Google DeepMind is facing a significant decline in its ability to attract and retain elite AI talent. According to analysis from Zeki Data, the lab’s share of research and advanced engineering hires in the Europe, Middle East, and Africa (EMEA) region fell from 49% during 2022–2023 to 18.6% for the 2025–2026 period.

While DeepMind remains a net recruiter, its arrival-to-departure ratio has dropped sharply from approximately 12:1 in Q2 2023 to 2:1 in Q3 2026. Competitors are aggressively absorbing DeepMind’s workforce; over the last year, 25% of departing researchers moved to Anthropic, 21% to Meta, and 14% to OpenAI.

High-profile departures include Nobel laureate John Jumper, who joined Anthropic, and several key researchers—Jeff Dean, Sanjay Ghemawat, Oriol Vinyals, and Quoc Le—who co-founded Discovery Loop. Former employees suggest the shift is driven by DeepMind’s changing focus toward the commercialization of the Gemini model, which some feel dilutes the lab’s original culture of open-ended, long-term scientific research.

While DeepMind continues to see growth in robotics and AI for science, it is losing ground in critical areas such as large language models (LLMs), multimodal AI, and computer vision. This talent shift occurs alongside a leadership restructuring where Demis Hassabis has transitioned to Alphabet Chief Scientist, leaving daily operations to Koray Kavukcuoglu.

Entities

Anthropic · Demis Hassabis · Google DeepMind · OpenAI · Zeki Data

Claims

What the coverage asserts, and how many sources carry each claim.