WeRide, a global autonomous driving technology company, has announced that its proprietary world model, WeRide GENESIS, has been named the “Overall Gen-AI Solution of the Year” at the 2026 AI Breakthrough Awards. The recognition places the company among leading technology innovators such as AMD, Qualcomm, and Dell Technologies.
The latest accolade comes just days after WeRide GENESIS received the Simulation Innovation Award at the 2026 Automotive Testing Technology International (ATTI) Awards on June 24, marking two major industry recognitions within 48 hours. The back-to-back awards highlight the platform’s growing role in accelerating autonomous driving development, simulation capabilities, and real-world deployment.
As artificial intelligence continues to evolve beyond digital content generation, industry attention is increasingly shifting toward Physical AI—systems capable of understanding and interacting with real-world environments. Unlike conventional generative AI models, Physical AI applications must comprehend physical laws, causal relationships, and spatiotemporal dynamics to make safe, reliable, and executable decisions in dynamic environments.
Within this emerging landscape, autonomous driving has become one of the most commercially viable and scalable applications of Physical AI. The sector uniquely combines large-scale real-world data collection with sustainable commercial operations, creating a powerful foundation for continuous AI learning and development.
WeRide believes that the ability to bridge digital intelligence with physical-world understanding will define the next phase of AI innovation. Through GENESIS, the company aims to accelerate the development of autonomous driving systems by enhancing simulation accuracy, training efficiency, and real-world scenario generation, helping advance the commercialization of intelligent mobility solutions.
As one of the autonomous driving companies with the broadest global operational footprint, WeRide is emerging as a builder of this foundational Physical AI infrastructure. Through its world model, WeRide GENESIS, the company connects real-world driving environments with high-fidelity simulation, accelerating large-scale development, training, and deployment of autonomous driving systems.
WERIDE GENESIS: FOUR-LAYER ARCHITECTURE
WeRide GENESIS is constructed on four integrated capability layers:
- Physical World Reconstruction: Generates pixel-level-fidelity road environments in minutes, including rare safety-critical long-tail scenarios that are impractical to capture through real-world data collection alone.
- Physicalized Traffic Interaction: Models complex traffic dynamics—pedestrian behavior, road surface conditions, occlusion risks, and collision boundaries.
- Causal Reasoning and Inference: Models cause-and-effect relationships across traffic participants, enabling robust decision-making in dynamic environments.
- Spatiotemporal Prediction: Simulates how scenarios evolve across space and time, generating probabilistic future outcomes to support driving strategy optimization.
Together, these capabilities enable autonomous driving systems to move beyond perception toward a deeper understanding of the physical world. By training and validating systems in large-scale, high-fidelity virtual environments, WeRide GENESIS can compress millions of kilometers of road testing into days while reducing data collection and annotation costs by more than 75%.
Ultimately, the true value of a world model must be proven in real-world deployment. Level 4 fully driverless operation represents the most rigorous and definitive proving ground for Physical AI capabilities.
WeRide is driving its global expansion with the WeRide GENESIS as its core engine. The company’s autonomous driving products have now been deployed in more than 40 cities across 12 countries, completing cross-city, cross-border, and cross-road-environment commercial validation. Currently, WeRide’s Robotaxi has commenced fully driverless commercial operations in Guangzhou and Beijing (China), as well as in Abu Dhabi and Dubai (UAE). Meanwhile, its services are open to the public in Singapore and Riyadh (Saudi Arabia), and fully driverless public operations will soon be launched in Madrid (Spain) and Zurich (Switzerland), among other locations.
At the same time, the world model capabilities of GENESIS are making a breakthrough in the L2++ mass-production assisted driving market
At the Second China Intelligent Driving Competition, the Chery Exeed Sterra ES equipped with WeRide’s one-stage end-to-end ADAS solution, WRD 3.0, secured a historic achievement—becoming the first and only vehicle in the competition’s history to win six consecutive championships.
Across a wide range of evaluation scenarios—including urban roads, highways, and parking environments—WRD 3.0 consistently demonstrated stable, smooth, and reliable driving performance. With a significantly more efficient R&D approach, the system outperformed competitors with multiple times the resources, fundamentally redefining the traditional development path built on scaling manpower, vehicle fleets, and rule-based systems.
This breakthrough marks what can be seen as the “ChatGPT 3.5 moment” for autonomous driving.
To date, WRD 3.0 has secured mass-production design wins across more than 30 vehicle models with leading OEMs including Chery and GAC Group. As these programs advance, the capabilities of WeRide GENESIS are rapidly translating into product competitiveness for the passenger vehicle market.
The universality of physical laws allows a world model trained on urban traffic to extend into a much broader range of real-world interaction scenarios. As more vehicles, cities, and environments are integrated into a unified model foundation, WeRide GENESIS continues to amplify data value and improve development efficiency, forming a self-reinforcing flywheel from data accumulation to model training, simulation validation, and real-world deployment.
Looking ahead, WeRide will continue to advance WeRide GENESIS as the foundational platform for Physical AI, enabling AI systems not only to understand the physical world, but to operate within it at scale—accelerating the global commercialization of autonomous driving technologies.
