×
Sweden’s Volvo creates AI virtual world to accelerate its zero-accident mission
Written by
Published on
Join our daily newsletter for breaking news, product launches and deals, research breakdowns, and other industry-leading AI coverage
Join Now

Volvo is leveraging an AI-generated virtual world to accelerate its zero-accident mission, a bold safety initiative aimed at preventing traffic fatalities. By creating high-fidelity digital environments that can be easily manipulated, the Swedish automaker is dramatically shortening the development cycle for its advanced driver assistance systems (ADAS). This technological breakthrough represents a significant advancement in how carmakers can test safety features by simulating countless dangerous scenarios without real-world risk.

The big picture: Volvo has partnered with Zenseact and the Wallenberg AI research program to create an AI-driven testing environment that transforms real-world driving data into adaptable virtual scenarios.

  • Using Gaussian splatting, an advanced rendering technique, engineers can visualize and manipulate realistic three-dimensional driving environments in days rather than months.
  • The technology allows Volvo to test its vehicle systems against rare but dangerous scenarios that would be impractical or impossible to recreate in physical testing.

Why this matters: The technology could dramatically accelerate Volvo’s progress toward its zero-accident goal as the World Health Organization estimates 1.19 million annual traffic-related deaths globally.

  • By multiplying limited real-world data into thousands of scenario variations, Volvo can train its AI systems on a much broader range of potential hazards.

How it works: Volvo collects real-world incident data through sensors on its connected vehicles, capturing emergency braking events, sharp steering maneuvers, and driver interventions when ADAS systems are operating.

  • Engineers can then reconstruct these incidents virtually and manipulate variables—adding or removing roads, changing traffic behaviors, or inserting different obstacles—to test how safety systems respond.
  • The technology can generate extreme hazards to explore vehicle behavior in complex but infrequent accident scenarios.

What they’re saying: “Data holds the key to improving automotive safety. Thanks to Gaussian splatting, we can now quickly multiply the millions of datapoints we have, turning real-world sensor sequences into thousands of variations of edge cases,” Alwin Bakkenes, head of global software engineering at Volvo Cars, told Newsweek.

  • Bakkenes described the technology as “the mileage multiplier that we’ve been looking for to accelerate how quickly we reach our ambition for a zero-collision future.”

Behind the technology: The virtual testing environment is powered by a newly expanded partnership with NVIDIA, utilizing their DGX systems to process and contextualize massive amounts of driving data.

  • Volvo and Zenseact have jointly established one of the largest data centers in the Nordic region to support this AI development work.
  • The company’s new software-defined vehicles, including the EX90 electric SUV and forthcoming ES90 sedan, are designed to receive over-the-air updates, allowing safety systems to continuously improve.

The complete approach: Volvo is combining its virtual environment testing with traditional real-world software training, development, and validation methods to create a comprehensive safety development ecosystem.

Volvo using AI, your SUV's data to create a virtual world to test safety

Recent News

AI agents reshape digital workplaces as Moveworks invests heavily

AI agents evolve from chatbots to task-completing digital coworkers as Moveworks launches comprehensive platform for enterprise-ready agent creation, integration, and deployment.

McGovern Institute at MIT celebrates a quarter century of brain science research

MIT's McGovern Institute marks 25 years of translating brain research into practical applications, from CRISPR gene therapy to neural-controlled prosthetics.

Agentic AI transforms hiring practices in recruitment industry

AI recruitment tools accelerate candidate matching and reduce bias, but require human oversight to ensure effective hiring decisions.