GOOGLE WROTE A ‘ROBOT CONSTITUTION’ TO MAKE SURE ITS NEW AI DROIDS WON’T KILL US

Summary from the TLDR newsletter

The DeepMind robotics team has revealed three new advances that will help robots make faster, better, and safer decisions in the wild. One involves a system for gathering training data with a set of safety-focused prompts instructing the model to avoid choosing tasks that involve humans, animals, sharp objects, and electrical appliances. The others are a neural network architecture designed to make the existing Robotic Transformer RT-2 more accurate and faster and RT-Trajectory, which adds 2D outlines to help robots better perform specific physical tasks.