Top AI Researchers Depart Meta, Google, OpenAI to Launch Billion-Dollar Startups

A wave of top AI researchers from major technology companies like Meta, Google, and OpenAI are leaving to start their own AI-focused startups. These fledgling companies are quickly securing enormous funding rounds, signaling strong investor confidence in their innovative approaches.

This trend highlights a shift in the AI industry, where smaller, agile startups are capitalizing on gaps left by large AI labs focused on immediate commercial goals. The movement could reshape the future of AI research and development.

Big Tech Talent Launches AI Startups with Record Funding

Former researchers from AI giants such as Google DeepMind and Meta are rapidly founding startups and raising unprecedented capital. For instance, David Silver, ex-Google DeepMind, secured a record $1.1 billion seed round for Ineffable Intelligence within months of its inception. Similarly, Tim Rocktäschel, another ex-DeepMind scientist, is reportedly raising up to $1 billion for Recursive Superintelligence.

AMI Labs, founded by Yann LeCun after departing Meta’s AI chief role, raised $1 billion to develop AI systems that learn continuously from real-world data. Other startups like Periodic Labs, Ricursive Intelligence, and Humans& have also attracted hundreds of millions in funding, often hiring talent from their founders’ former employers.

Why Are Researchers Leaving Big AI Labs?

Investors and experts suggest that the intense race among large AI labs to dominate the market has narrowed research focus, sidelining promising areas like new architectures and interpretability. Elise Stern, managing director at Eurazeo, explains that this focus creates a vacuum, opening opportunities for startups to explore neglected AI domains.

Alexander Joël-Carbonell of HV Capital notes that the pressure on big labs to deliver rapid benchmark results limits exploratory research, especially beyond the dominant large language model (LLM) paradigm. This environment motivates researchers to seek freedom and innovation in startups.

Startups Targeting Untapped AI Applications

Ricursive Intelligence, founded by former Anthropic and DeepMind staff, is developing AI tools for chip design automation, raising $335 million shortly after launching. Its founders emphasize neutrality to gain trust from chipmakers, a position difficult to maintain within large tech firms.

Periodic Labs, backed by ex-OpenAI and DeepMind talent, raised $300 million to build autonomous labs. AMI Labs focuses on overcoming AI’s challenges with grounding, causality, and real-world reliability, critical as AI expands into robotics, healthcare, and industry.

The Future of AI Innovation: Startups vs. Big Labs

With venture capital pouring $18.8 billion into AI startups since early 2025, surpassing last year’s pace, the AI ecosystem is rapidly evolving. Founders with insider knowledge of large labs are uniquely positioned to identify and exploit overlooked opportunities.

As AI moves beyond traditional applications, startups focusing on reinforcement learning and novel AI architectures may drive the next wave of breakthroughs. This shift could redefine AI research, balancing commercial pressures with exploratory innovation.

"When you're in a race, you narrow focus. That creates a vacuum. Entire areas of research, like new architectures, agents, interpretability and vertical models, are being deprioritized, not because they don't matter, but because they don't win the immediate race."—Elise Stern, Managing Director at Eurazeo