Amazon’s Zoox Accelerates Robotaxi Race With U.S. Expansion

Zoox confirmed a significant expansion of its autonomous vehicle testing program into Phoenix and Dallas. The move increases the company’s testing footprint to 10 U.S. metropolitan areas and represents one of the largest geographic expansions since Amazon acquired Zoox for approximately $1.2 billion in 2020.

The expansion is strategically timed as competition intensifies across the autonomous ride-hailing sector. Zoox is attempting to close the operational gap with Waymo, which currently operates the largest commercial robotaxi network in the United States, and Tesla, which began limited driverless ride services in Austin in early 2026.

According to industry data compiled by mobility analysts, autonomous vehicle testing miles across major U.S. developers increased by over 45% between 2024 and 2025, driven largely by the rapid improvement of AI perception models and onboard computing systems.

Why Phoenix and Dallas Were Selected

Phoenix: Extreme Heat and High-Speed Network Testing

Phoenix has already become one of the most important testbeds for autonomous mobility in the United States. The region’s weather and road infrastructure create a unique technical environment for autonomous driving systems.

Summer temperatures in Phoenix routinely exceed 110°F (43°C), placing stress on vehicle electronics, battery thermal management systems, and sensor reliability. For autonomous vehicles, high temperatures can affect LiDAR accuracy, camera sensor calibration, and computing system performance.

Phoenix also presents another challenge: large-scale suburban road infrastructure. The city’s metropolitan area spans more than 14,500 square miles, making it one of the largest urban footprints in the country. Roads frequently feature six-lane arterial corridors and speed limits reaching 55–65 mph, conditions that differ substantially from the lower-speed urban grids where many early autonomous vehicle tests took place.

Zoox engineers are using Phoenix to validate how the company’s autonomous system performs under these conditions. Testing includes evaluating sensor accuracy under intense sunlight, analyzing vehicle behavior at higher speeds, and validating navigation algorithms across sprawling suburban road networks.

Dallas: Infrastructure Complexity and Weather Variability

While Phoenix provides environmental stress testing, Dallas introduces a different set of technical variables.

The Dallas–Fort Worth metroplex contains over 75,000 lane-miles of roadway, including some of the most complex highway systems in the United States. Multi-level freeway interchanges, multi-lane merges, and heavy commuter traffic patterns create scenarios that challenge autonomous planning systems.

In addition to infrastructure complexity, Dallas experiences more variable weather than Phoenix. Sudden thunderstorms, heavy rainfall, and visibility changes require autonomous systems to adapt rapidly to changing road conditions.

For Zoox, this environment allows its AI perception and planning systems to be tested under conditions that combine dense infrastructure with environmental unpredictability.

Zoox’s Expansion Strategy: Mapping First, Autonomy Later

Zoox is implementing a phased rollout model that it has used in previous markets. The initial stage involves manual mapping operations.

During this phase, vehicles based on the Toyota Highlander platform are deployed across city streets. These vehicles are equipped with advanced sensor arrays that collect detailed environmental data including lane markings, traffic signals, pedestrian infrastructure, and curb locations.

The mapping process produces high-definition three-dimensional maps with centimeter-level accuracy, which autonomous driving systems use for localization and navigation.

Once mapping is complete, Zoox begins supervised autonomous driving tests. Vehicles operate using the autonomous system but retain a trained safety driver behind the wheel.

This step allows engineers to validate performance under real-world conditions before removing the human operator.

The Purpose-Built Zoox Robotaxi

Unlike many competitors that retrofit existing vehicles, Zoox has developed a fully custom autonomous vehicle architecture.

The Zoox robotaxi features a symmetrical design that allows the vehicle to move forward or backward without turning around. The vehicle does not include a steering wheel or traditional driver controls, reflecting its design for fully autonomous operation.

Passenger seating is arranged in a carriage-style configuration where riders face each other rather than facing forward. This layout is intended to improve passenger comfort and maximize interior space for shared mobility services.

The vehicle is powered by a battery system designed for urban ride-hailing operations, with an estimated range of up to 16 hours of operation on a single charge under urban driving conditions, according to earlier engineering disclosures.

A Critical Phase in the Robotaxi Race

Zoox’s latest expansion demonstrates how rapidly the autonomous mobility race is evolving.

With Amazon’s financial backing and growing infrastructure investments, Zoox is positioning itself as a serious competitor in the emerging robotaxi market.

The next phase of development will likely determine whether the company can transition from limited testing programs to large-scale commercial deployment.

If successful, autonomous ride-hailing networks could fundamentally reshape urban transportation in the coming decade.