Generalist valued at $3B in expanded robotics funding round

Generalist, an artificial intelligence robotics company, has climbed to a $3 billion valuation following a capital injection of approximately $200 million. The investment, led by 8VC, extends the company's Series B round initially announced in June at a $2 billion valuation, bringing total Series B funding to $600 million—a contribution that included backing from Radical Ventures.

Building a Universal Robot Brain#

The startup, which launched in 2024, aims to develop a foundational AI model capable of operating across different robotic platforms. The company's latest release, the Gen 1.5 model, claims to enable robots to learn new behaviors from brief video clips—as little as 3 to 12 seconds of demonstration. This approach represents a departure from traditional robotics programming, where machines require explicit instruction for each distinct task.

Generalist was founded by Pete Florence and Andy Zeng, both former researchers at Google DeepMind, alongside Andrew Barry, who previously worked at Boston Dynamics. The company has secured backing from a constellation of prominent investors, including 8VC, Radical Ventures, Nvidia, Union Square Ventures, Bezos Expeditions, and AI researcher Fei-Fei Li.

Navigating a Crowded Field#

The robotics sector has become increasingly competitive, with multiple startups pursuing similar ambitions. Physical Intelligence commands a reported $11 billion valuation, while Skild AI, backed by SoftBank, sits at $14 billion. Genesis AI was reportedly exploring a $3 billion valuation as recently as last month. Despite facing well-funded competitors, Generalist has maintained a relatively low profile until recent months.

The ChatGPT Question#

The funding wave underscores investor optimism about robotics reaching what some call a "ChatGPT moment"—a threshold where machines could execute broad tasks without bespoke training. Yet venture capitalists acknowledge a significant limitation: unlike large language models trained on internet-scale data, robots cannot easily be trained at comparable scale, suggesting that truly general-purpose robotic systems may still require years of development.