A vision-language-action (VLA) model is an artificial intelligence system that combines visual perception, natural-language understanding, and motor control to enable robots or agents to see, interpret, and physically interact with their environment. By integrating these three domains, a VLA model allows a robot to understand spoken or written instructions, identify relevant objects, plan a sequence of actions, and execute them autonomously.
Shift, the new startup that’s offering to clean NYC apartments for free in exchange for in-home video data to train robots, is based in Germany, where it would face much tougher legal barriers.
The Huawei-backed startup GigaAI just launched its AI-powered robot maid, the SeeLight S1.
Beijing’s Galbot just reached a new level of global visibility thanks to a viral visit to one of its futuristic storefronts during Trump’s summit with President Xi.
Several Massachusetts startups are generating global buzz as the state aims to reclaim its position as a leader in physical AI.
Generalist AI has been publishing research since January 2026, but their latest release has generated a lot of buzz with their GEN-1 model.
One of Tesla’s biggest rivals in the humanoid race just got the attention of Elon Musk with its latest demo.
Xiaomi says it’s moved its humanoid robot into its already highly automated EV assembly line to take on the remaining tasks that still require human workers.
Figure AI says its flagship humanoid is more autonomous than ever thanks to its newly upgraded robot brain.
China just open-sourced a robot brain the government says can safely think, move, and work on its own.
The rising Shenzhen startup LimX Dynamics says its new robot, the TRON 2, is designed to collect the kind of data that’s considered most valuable for training humanoids to take over human jobs.
The latest paper from the $5.6B embodied AI startup Physical Intelligence is likely to intensify the growing race to collect raw video data to train next-generation robots to take over human jobs.