Rodney Brooks, a pioneering roboticist and co-founder of iRobot, has issued a stark warning to investors pouring billions into humanoid robotics: the current industry trend is a financial dead end. According to Brooks, the prevailing strategy of attempting to teach robots dexterity by imitating human behavior through video analysis is nothing more than “pure fantasy thinking.”
The Fallacy of Humanoid Dexterity
Brooks argues that the industry fundamentally underestimates the complexity of human touch. Unlike speech recognition or image processing, which benefited from decades of structured data collection, the field of touch-based robotics lacks a comparable foundation. Human hands are equipped with approximately 17,000 specialized touch receptors, a level of sensory sophistication that current robotic hardware cannot replicate. In his recent essay, Brooks details why these biological limitations remain a massive roadblock for startups like Tesla and Figure.
Physics and the Safety Paradox
Beyond the software hurdles, physics presents a dangerous reality for full-sized humanoid machines. These robots require immense amounts of energy to maintain stability while walking. Brooks highlights that a robot double the size of current models would possess eight times the kinetic energy, making potential falls catastrophic in human-centric environments. He predicts that within 15 years, the most successful robots will abandon the human form entirely in favor of wheeled designs, multiple arms, and specialized sensors optimized for specific tasks.
The AI Productivity Mirage
Brooks’ skepticism extends to the broader hype surrounding generative AI. He points to research from the nonprofit METR, which examined the impact of AI tools on professional software developers. The study revealed a counterintuitive result: developers using AI took 19% longer to complete tasks, despite 20% of them perceiving that the tools had actually increased their speed. This gap between perceived efficiency and actual output mirrors his concerns regarding the overvaluation of current robotics startups.
Big Tech and the Funding Frenzy
Despite these technological and physical constraints, major corporations continue to exert a powerful gravitational pull on the sector:
- Apptronik: Having secured nearly $450 million, the firm maintains a strategic partnership with Google’s DeepMind to merge AI with embodied hardware.
- Figure: Recently announced a massive funding round, with a valuation reaching an astonishing $39 billion. This follows a high-profile partnership with OpenAI, which the company ended in March after claiming a “major breakthrough” in its proprietary, end-to-end robotics AI.
Brooks remains unconvinced that these massive capital injections will lead to scalable mass production. Having spent decades observing the evolution of machine learning and the rise of data-centric companies, he maintains that the current path of trying to force robots into human shapes is a misguided bet that ignores the realities of both engineering and physical limitations.