GM Scales Up AV Test Fleet
GM has rolled out its 100th autonomy test vehicle equipped with production-intent hardware. The company is scaling its multi-modal data collection effort faster than all of 2025 combined, though some critics are dismissing the hardware-heavy approach as "Cruise 2.0" compared to Tesla's vision-only strategy.
GM's new AV strategy represents a significant pivot, moving away from the robotaxi model after shuttering its Cruise subsidiary in late 2024. The company is now focusing on developing personally owned autonomous vehicles by integrating the former Cruise engineering team with its Super Cruise advanced driver-assistance system (ADAS) team. This move follows a tumultuous period for Cruise, which faced regulatory scrutiny and suspended its driverless operations after a pedestrian-dragging incident in 2023. The new test fleet is composed of Cadillac Escalade IQs and GMC Yukons, a shift from the modified Chevrolet Bolts used by Cruise. These vehicles are equipped with a multi-modal sensor suite that includes a combination of cameras, short- and long-range radars, and lidar, enabling a 360-degree, three-dimensional view of the vehicle's surroundings. This hardware-intensive approach is what draws comparisons to the original Cruise strategy, as it contrasts sharply with Tesla's reliance on cameras alone. The immediate goal for this new fleet is to gather extensive road data to develop a Level 3 "eyes-off" driving system, which GM aims to launch on the Cadillac Escalade IQ in 2028. This system is intended to allow the driver to safely disengage from monitoring the road under certain highway conditions. The technology is being engineered to handle 95% of all driving scenarios on paved public roads in the U.S. and Canada over time. Underpinning this ambition is a new centralized computing architecture, also slated for a 2028 debut. This platform will consolidate systems like propulsion, safety, and infotainment onto a single computer, designed to provide up to 35 times more AI performance and significantly increased bandwidth for over-the-air updates. This architecture will be scalable across both electric and internal combustion vehicles.
