AI² Robotics Founder Defends VLA, Launches NeuroVLA Model

Guo Yandong, founder and CEO of AI² Robotics, stated that Vision-Language-Action (VLA) architectures are central to embodied AI. He emphasized that perception, reasoning, and action are inseparable in physical intelligence systems. The newly introduced NeuroVLA model features a brain-inspired architecture that reduces response latency from hundreds of milliseconds to tens of milliseconds. Additionally, the AlphaBrain Platform was launched as an open-source toolkit that supports brain-inspired VLA models and major benchmarks like LIBERO and CALVIN.