AI news ·
Meta acquires robotics AI startup to accelerate humanoid robot development
Meta acquired robotics AI startup Assured Robot Intelligence to accelerate humanoid robot development. The team will join Meta's Superintelligence Labs, but financial terms were not disclosed.
Meta Platforms has acquired Assured Robot Intelligence (ARI), a startup building AI models that help robots understand and adapt to human behavior in complex environments. The deal, confirmed by a Meta spokesperson to Bloomberg, accelerates the company's push into humanoid robotics through its newly formed Meta Robotics Studio.
Financial terms were not disclosed. The ARI team, including co-founders Lerrel Pinto and Xiaolong Wang, will join Meta Superintelligence Labs' research division and work closely with the Robotics Studio team launched last year to develop underlying humanoid technology.
What ARI brings to Meta
Meta said the startup is "at the frontier of robotic intelligence designed to enable robots to understand, predict and adapt to human behaviors in complex and dynamic environments." The company's website describes its work as building frontier AI that helps robots address critical challenges in high-value labor markets.
Wang previously worked as a researcher at Nvidia. Pinto co-founded Fauna Robotics before leaving in 2025. Amazon acquired Fauna in March to strengthen its own humanoid robot efforts, Bloomberg reported.
The competitive landscape
Several major tech companies have accelerated humanoid robot development this year. Tesla showcased Optimus Gen 3 in Shanghai at the Appliance & Electronics World Expo in early 2026. The robot uses the same "AI5" inference chips found in Tesla's latest vehicles, with a full reveal targeted for late 2026.
Google announced a partnership with Agile Robots SE in March to deploy reasoning-capable robots in industrial manufacturing settings. The moves signal intensifying competition to commercialize humanoid robots for labor-intensive industries.
Why this matters for IT and development professionals
The acquisition signals growing demand for engineers who can bridge AI for IT & Development with physical systems. Humanoid robots require software stacks that integrate perception models, real-time inference, and behavioral prediction - skills that overlap heavily with modern AI engineering. Developers working on embodied AI or robotics middleware will find these skill sets increasingly transferable to large-scale commercial projects. For those building expertise in this area, an AI Learning Path for Software Developers provides structured grounding in the model development and deployment techniques that underpin systems like those ARI builds.