Sterling Anderson Doesn't Think Humanoid Robots Should Be Building Cars

General Motors' product chief breaks down the company's software evolution, the physics of multisensor vehicle safety, and why personally owned autonomous vehicles beat robotaxis.

By Andrew J. Hawkins

GM's Software Evolution

At General Motors, Sterling Anderson has taken on a central role in reshaping how the company approaches software in its vehicles. As GM's product chief, Anderson is responsible for driving a shift toward a more software-defined vehicle architecture — a transition that legacy automakers have struggled to execute at the pace of newer entrants.

Anderson's perspective on that evolution is grounded in a practical view of what the vehicle of the near future should be: increasingly capable, continuously updateable, and built on a foundation of rigorous safety engineering rather than novelty.

The Case for Multisensor Safety

A key part of Anderson's argument centers on the physics of vehicle safety. He emphasizes that no single sensor modality is sufficient for robust perception, and that combining multiple sensor types — cameras, radar, and lidar — provides the kind of redundancy that safety-critical systems demand. Each technology has physical limitations, and layering them together allows a vehicle to compensate when one modality is degraded by weather, lighting, or other conditions.

Why Personally Owned Autonomous Vehicles Beat Robotaxis

Anderson is also outspoken on the future of autonomy: he favors personally owned autonomous vehicles over the robotaxi model. In his view, putting autonomous capability directly into vehicles that people buy and use every day delivers more value — and more safety — than concentrating autonomy in a centralized fleet serving limited geographies.

Humanoid Robots and the Future of Manufacturing

As for humanoid robots building cars? Anderson is not convinced. His skepticism reflects a broader debate in the industry about where humanoid robotics genuinely adds value versus where existing automation — purpose-built industrial robotics — remains more effective.

Topics: Transportation, Autonomous Cars, Electric Cars

via The Verge