Unitree's Market Surge, Thrun's New Venture, and AI in Turbulence Control

Unitree Robotics' shares soared 629% on debut, Sebastian Thrun launches Dulo, and HydroGym trains AI for turbulence control.

Unitree Robotics' Explosive Market Debut

Unitree Robotics, a leading Chinese robotics company, experienced a remarkable surge in its stock market debut on the Shanghai Star Market, closing at 845 yuan after an initial offering of 150.80 yuan. This represents a staggering 460% increase, highlighting the intense investor interest in humanoid robotics. Founded in 2016, Unitree has become a key player in China's robotics ambitions, shipping over 5,500 humanoid robots last year and achieving a net profit of 278 million yuan in 2025.

The company's success can be attributed to its strategic use of China's robust tech supply chain and government backing. Unitree focuses on producing cost-effective humanoid robots, such as its $13,500 G1 model, which compete with pricier Western counterparts. The firm's technology includes advanced sensors, automated arms, and human-like machines, positioning it as a leader in the global robotics market.

Unitree's market debut underscores the growing demand for robotics technology, driven by China's strategic prioritization of AI and robotics. With 97% of global humanoid robot shipments in the first half of 2026 coming from Chinese companies, Unitree's success reflects a broader trend of Chinese dominance in this sector. This positions Unitree as a formidable competitor to US firms like Tesla and Boston Dynamics.

From my perspective, Unitree's rise is a testament to the potential of integrating government support with cutting-edge technology. The company's ability to deliver high-quality robots at competitive prices could reshape the global robotics landscape, particularly as demand for automation grows in response to labor shortages and economic pressures.

Sebastian Thrun's Stealth Robotics Startup: Dulo

Sebastian Thrun, renowned for pioneering self-driving cars, has launched a new robotics startup named Dulo, currently operating in stealth mode. Announced at the Actuate robotics conference, Thrun revealed that Dulo aims to revolutionize hardware design through 'foundation models,' enabling rapid manufacturing processes. The startup's team includes experts from Waymo, Google Brain, and Stanford's AI Laboratory, indicating a strong foundation in AI and autonomy.

Dulo's focus on applying foundation models to hardware design marks a departure from typical robotics startups, which often concentrate on building robots themselves. Instead, Dulo seeks to enhance the design and manufacturing process, potentially offering tools to other companies in the robotics industry. This approach could address the slow pace of physical design compared to software advancements.

The implications of Dulo's venture are significant, as it enters a booming robotics market characterized by falling hardware costs and increased investment. With physical AI startups raising $16.3 billion in early 2026, Dulo's innovative approach could attract substantial funding, leveraging Thrun's reputation and expertise. The company's success could drive further advancements in manufacturing efficiency, benefiting the broader robotics industry.

In my view, Thrun's initiative with Dulo represents a strategic shift towards optimizing the foundational aspects of robotics. By focusing on design and manufacturing, Dulo could accelerate the development of next-generation robotics, aligning with trends in AI-driven automation and reshoring manufacturing capabilities.

HydroGym: AI Tackling Turbulence Control

HydroGym, a platform developed by Lagemann et al. and reported in Nature, is making waves by training AI models to control turbulent fluid flows. Turbulence, a longstanding challenge in physics and engineering, affects everything from aviation to wind turbine efficiency. HydroGym employs reinforcement learning, allowing AI to learn control strategies through trial and error, which can then be transferred across different fluid environments.

The technical backbone of HydroGym involves using AI to simulate fluid dynamics and optimize control strategies. By leveraging reinforcement learning, AI models can adapt to complex and unpredictable turbulence patterns, potentially offering solutions that traditional methods cannot. This innovative approach could revolutionize how industries manage fluid dynamics, leading to more efficient systems.

The potential impact of HydroGym extends beyond fluid dynamics, as it could influence various sectors reliant on fluid flow control. If experimentally verified, these AI-driven strategies could enhance efficiency in industries like aerospace and energy, offering a competitive edge. The platform's success could also spur further research into AI applications in other complex physical systems.

Editorially, HydroGym exemplifies the transformative power of AI in tackling intricate scientific problems. By harnessing AI's adaptability and learning capabilities, HydroGym could set a precedent for future innovations in engineering and beyond. This aligns with broader trends of integrating AI into traditional industries to enhance performance and sustainability.

AI News Center