PlusAI has released the latest version of its autonomous trucking software platform. SuperDrive 6.0 introduces new capabilities designed to support commercial-scale driverless freight operations, including night driving and the ability to navigate construction zones.

The Santa Clara, California–based company said the update significantly accelerates development of its autonomous system while expanding the environments in which its trucks can operate.

The company is targeting fully driverless commercial deployment with factory-built autonomous trucks in 2027.

“SuperDrive 6.0 isn’t an incremental update; it’s a major advancement of what an autonomous ‘brain’ can do,” said David Liu, CEO and co-founder of PlusAI. “By adding night driving and construction zone handling, autonomous trucks with SuperDrive could achieve 24/7 commercial operations.”

Expanding Operating Conditions

The new software introduces two capabilities considered essential for large-scale freight operations:

• Night driving
• Construction-zone handling

According to PlusAI, trucks equipped with SuperDrive 6.0 are already hauling freight on commercial routes in Texas, where the construction-zone feature is active.

The company expects to enable night driving in the coming weeks, which would allow trucks to operate continuously across customer routes.

Operating overnight could significantly increase equipment utilization. The company noted that 24/7 operations could potentially more than double the productivity of a truck compared with typical daytime operations.

Construction zones are also a key safety challenge for trucking. Thousands of crashes occur in Texas work zones each year, often tied to speeding, driver distraction or unsafe lane changes.

Faster AI Development

A central element of the release is a major improvement in the speed and cost of training the autonomous driving system’s artificial intelligence models.

PlusAI said SuperDrive 6.0 delivers:

• 10-times faster AI model training
• 3-times reduction in data-labeling costs

The improvements stem from a combination of auto-labeling, imitation learning and reinforcement learning techniques that accelerate how quickly the system can learn from real-world and simulated driving scenarios.

According to the company, the faster development cycle allows new features to move from internal validation to commercial deployment in a matter of weeks. This enables quicker expansion into new routes and operating environments.

Production-Focused Architecture

SuperDrive 6.0 is built on what PlusAI describes as a production-grade distributed computer architecture designed for commercial trucking operations.

The system runs across multiple high-performance system-on-chips, including Nvidia Drive Orin and Thor, allowing the autonomous system to maintain performance even when sensors degrade or hardware components experience partial failures.

At the software level, the system incorporates a new Transformer-based “Reflex” layer within the company’s AV 2.0 architecture.

The layer integrates large-scale perception with motion forecasting to better predict how surrounding vehicles, pedestrians and other road users will move.

PlusAI said the motion forecasting model delivers a twofold improvement in predicting the trajectories of dynamic road users, enabling safer and smoother driving decisions.

Built on Global Real-World Data

The software builds on more than 7 million miles of real-world driving across the United States, Europe and Asia.

PlusAI said the latest update represents a key milestone toward scaling autonomous freight operations.

The company is pursuing commercialization alongside partnerships with truck manufacturers and plans to deploy factory-built autonomous trucks as part of its driverless launch strategy.

The announcement comes as PlusAI prepares to go public through a business combination with Churchill Capital Corp IX.

With SuperDrive 6.0 now operating in commercial freight routes, the company says the new platform positions it to expand autonomous trucking operations while continuing to refine the technology ahead of its planned 2027 driverless commercial launch.