Boston Dynamics partnered with Google Cloud and DeepMind to integrate Gemini Robotics-ER 1.6 into its Spot robot dog and Orbit inspection platform, giving a robot that used to run scripted or teleoperated routines a general-purpose embodied-reasoning model for spatial understanding and autonomous decision-making. The approach reflects a pattern spreading across robotics in 2026: rather than each hardware maker building its own bespoke perception-and-planning stack from scratch, they license or integrate a general embodied-reasoning model as the "brain," while keeping their own hardware, actuators, and low-level control as the "body." Gemini Robotics-ER is built specifically to reason about physical space and plan movement, so bolting it onto Spot's already-mature hardware produces a machine that can inspect a facility, identify anomalies, and decide what to do next without a human driving it frame by frame, instead of just executing pre-programmed patrol routes. This mirrors what is happening at Tesla with Optimus, at Figure AI, and across the humanoid field: hardware reliability was mostly solved years ago, and the current bottleneck is the reasoning layer that lets a robot handle situations nobody explicitly programmed for. The practical implication for builders in this space is that the economics of robotics are starting to look more like the economics of software, where you assemble a stack from a general-purpose foundation model plus specialized hardware, rather than training bespoke perception systems in-house. The honest caveat is that navigation and reasoning are still being integrated somewhat separately, and a controlled demo inside a Boston Dynamics facility is a long way from unattended reliability on a real industrial night shift, so this pattern is promising but not yet proven at the reliability bar that industrial deployment actually requires.