Waymo, the self-driving car company under Alphabet, has issued a recall for 672 of its vehicles following a crash in Phoenix, Arizona, where a robotaxi drove into a creek. The incident, which occurred on March 4, 2024, involved a vehicle that entered a construction zone and struck a pole before ending up in a shallow waterway. No injuries were reported, but the event has prompted immediate action from Waymo and scrutiny from British regulators.
The recall, filed with the US National Highway Traffic Safety Administration (NHTSA), covers all 672 vehicles in its current fleet. Waymo stated that the issue stems from a software error in its perception system, which failed to correctly predict the trajectory of a construction vehicle blocking a lane. The company has since deployed an over-the-air software update to all affected vehicles. This is the third recall for Waymo in 2024 alone, following two previous episodes related to sensor calibration and map data.
British regulators are watching closely. The Department for Transport and the Centre for Connected and Autonomous Vehicles have been monitoring Waymo's operations, as the company has been conducting trials in the UK. A spokesperson for the DfT said, 'We are in contact with Waymo and NHTSA to understand the root cause. Safety is our absolute priority, and we will ensure any lessons are applied to UK trials.' The UK has its own autonomous vehicle bill progressing through Parliament, which aims to establish a regulatory framework for self-driving cars by 2025.
The crash highlights the challenges of deploying autonomous vehicles in unpredictable environments. Waymo CEO Dmitri Dolgov said in a statement, 'We take full responsibility for this incident. Our system was tested against thousands of edge cases, but this one slipped through. We have already implemented a fix and are evaluating our testing protocols.' The recall comes at a time when public trust in autonomous vehicles is fragile. A recent survey by the UK's Thatcham Research found that 73% of British drivers would not trust a self-driving car, with safety concerns being the primary reason.
For British readers, the implications are clear. As the UK moves towards wider adoption of autonomous vehicle technology, incidents like this serve as a stark reminder of the need for robust regulation and oversight. The new UK legislation will require companies to prove their vehicles are 'safe enough' before deployment, but critics argue that definitions of 'safe enough' remain vague. The Waymo recall may accelerate calls for more rigorous testing and transparency.
From a tech perspective, the incident raises questions about the limits of current AI. Autonomous vehicles rely on deep learning models trained from massive datasets, but they struggle with 'out-of-distribution' scenarios, such as the unexpected presence of construction equipment. As one engineer at a rival firm noted, 'We can't train for everything. The real world is messy. The only solution is redundancy.' Indeed, Waymo's vehicles have multiple sensor types — lidar, radar, camera — but the software's prediction algorithm failed to integrate these inputs correctly.
For the average user, this recall is not a cause for alarm but a sign of an industry maturing. Recalls are common in traditional automotive manufacturing — BMW recalled 1.5 million vehicles just last year — and they represent a commitment to safety. The fact that Waymo caught the error before any injuries occurred is a testament to its monitoring systems. Yet, the frequency of recalls suggests that the path to fully autonomous driving is longer than some optimists predicted.
As the story develops, expect US regulators to increase their oversight, and UK regulators to embed new safety clauses in upcoming legislation. The creek crash may not be a major disaster, but it is a ripple that could reshape the landscape of autonomous transport.
