Noam Shazeer, a key figure in Google's AI efforts and co-lead of Gemini, is leaving the company to join OpenAI. His departure comes less than two years after Google reportedly paid $2.7 billion to bring him back.
Google's VP of Engineering and co-lead of the Gemini AI model, Noam Shazeer, is leaving Google to join OpenAI. This move highlights the intensifying competition for top AI talent among tech giants.
Meta CEO Mark Zuckerberg believes significant AI progress can be achieved with small, highly capable teams, contrasting with the industry's intense talent war. He highlighted this during a podcast, also discussing his nonprofit Biohub's work in AI and biology, and the challenges of compute resources.
Nvidia is aggressively competing for AI talent, offering top engineering roles base salaries up to ₹4.7 crore ($425,500). This, combined with substantial stock-based wealth, positions Nvidia as a top global tech employer, turning many employees into millionaires.
The AI industry is experiencing a talent war, with companies like OpenAI, Meta, and Google DeepMind competing for a few hundred elite engineers. Nine-figure compensation packages and massive equity grants are common, highlighting the immense value placed on individuals like Ilya Sutskever, Mira Murati, Alexandr Wang, Demis Hassabis, and Andrej Karpathy.
Over 600 OpenAI employees sold $6.6 billion in shares in October 2025, driven by intense AI talent competition, particularly after Meta's $300 million compensation packages for AI researchers, leading to increased salaries and stock-based compensation at OpenAI.
AI firms like OpenAI and Anthropic are poaching experienced sales and go-to-market executives from major software companies like Salesforce, Snowflake, and Datadog to drive enterprise growth. This trend highlights a shift in the AI talent war beyond researchers, as these companies increasingly focus on monetizing their technology through corporate clients.
Weiyao Wang, an eight-year Meta veteran who worked on multimodal perception systems and SAM3D, has joined Thinking Machines Lab following his final day at Meta last week. The $12B AI startup recently signed a multibillion-dollar cloud deal with Google for Nvidia's GB300 chips, placing it in the same infrastructure tier as Anthropic and Meta. The move comes as talent flows bidirectionally — Meta has poached seven of TML's founding members, while TML has hired more researchers from Meta than any other employer, including CTO Soumith Chintala (co-founder of PyTorch) and Piotr Dollár (co-author of Segment Anything).
A fierce talent war is underway in the AI sector, with defense tech and physical AI companies luring skilled engineers from self-driving vehicle companies. This is driving up salaries and creating a talent exodus from automakers and AV startups.
