Waabi is deploying its Level 4 autonomous truck system using a verifiable end-to-end AI architecture, a departure from the black box approaches used in passenger car Level 2+ systems.1 CEO Raquel Urtasun said those passenger car architectures "are not verifiable," positioning Waabi Driver as a more transparent alternative for commercial trucking.
The technology still faces weather limitations. Snowstorms remain a no-go zone for Waabi Driver.2 Urtasun acknowledged this gap while emphasizing the system's progress toward full autonomy in most conditions.
On workforce impact, Urtasun offered assurance to current drivers. "Everybody who's a truck driver today, and wants to retire as a truck driver, will be able to do so," she said.3 This timeline suggests autonomous trucks will phase in gradually rather than disrupt employment overnight.
The deployment comes as computer vision technology expands across automotive, retail, and security sectors. Vision-language-action models are enabling systems that combine visual perception with decision-making, moving beyond simple image recognition to autonomous operation.
Urtasun framed autonomous vehicles as a safety imperative. She noted 2 million deaths occur on roads globally per year, adding "nobody's questioning that."4 The comment positioned autonomous systems as a solution to human error rather than an experimental risk.
Meanwhile, Meta's Yann LeCun questioned the concentration of AI governance. "I don't think any of us, whether it's me or Dario [Amodei], Sam Altman, or Elon Musk, has any legitimacy to decide for society what is a good or bad use of AI," he said.1 The remark highlights tensions as autonomous systems move from research to real-world deployment.
Waabi's verifiable architecture represents a technical bet that transparency matters for Level 4 autonomy. Whether regulators and customers demand this over faster-to-market black box systems will determine if the approach becomes standard.