In a new interview, the Meta Chief AI Scientist and AMI Labs founder doubles down on world models, arguing that the massive data requirements of current humanoid demos prove the Vision-Language-Action (VLA) approach is a dead end.
Figure's unedited autonomous livestream has crossed 61 hours, sorting over 80,000 packages. CEO Brett Adcock addresses teleoperation skepticism, details a total shift away from Chinese supply chains, and teases Figure 4 as an "iPhone 1 moment" for robotics.
Less than two months after its massive Series A, RJ Scaringe’s Mind Robotics has secured an additional $400 million to scale its "Physical AI" platform within the automotive sector and beyond.
What began as a one-shift challenge has evolved into a multi-day endurance test for Figure’s Helix-02 neural network, as the company’s "Bob, Frank, and Gary" units attempt to hold the line under the gaze of thousands of live viewers.
Figure AI has launched its promised 8-hour livestream. While the robot is crushing its sub-3-second speed targets, some viewers are questioning the narrow scope of the demonstration.