Uber’s self-driving fleet has already run into trouble after one of the vehicles was spotted running a red light in San Francisco on 14 December. The incident has officially been blamed on “human error”.
A video of the incident was uploaded to YouTube by a taxi company after one of its dashboard cameras recorded it. It was first spotted by the San Francisco Examiner, which confirmed the details with the Luxor Cab company’s operations manager, Charles Rotter. The video shows a vehicle passing through a red light as a pedestrian has already begun to cross the road. The same thing apparently occurred in another area of the city the same morning, as spotted and tweeted about by a writer, Annie Gaus. She told the Examiner the Uber almost hit the Lyft car she was in: “It was close enough that both myself and the driver reacted and were like, ‘Shit’. It stopped suddenly and stayed like that.”
Uber, which had only just begun trialling cars in the city the day of the incident, has refused to obtain permits for testing the cars in the state of California, arguing there is no need because human drivers will always be present for the test.
In a blog announcing the San Francisco trials, posted the same day as the incident, Uber said: “We understand there is a debate over whether or not we need a testing permit to launch self-driving Ubers in San Francisco. We have looked at this issue carefully and we don’t believe we do. Before you think, 'there they go again' let us take a moment to explain...the rules apply to cars that can drive without someone controlling or monitoring them. For us, it’s still early days and our cars are not yet ready to drive without a person monitoring them.”
Read more: Uber acquires AI firm to boost self-driving car project
Now, however, it is blaming the red light incident on the very person meant to be doing said monitoring. “These incidents were due to human error,” a spokesperson told the Examiner. “This is why we believe so much in making the roads safer by building self-driving Ubers. These vehicles were not part of the pilot and were not carrying customers. The drivers involved have been suspended while we continue to investigate.”
WIRED has contacted Uber for further comment, and to find out whether any cars will be taken off the road. It’s unclear whether the human driver was operating the vehicle, or should have overridden the computer system.
It’s a fairly moot point, saying the car was not carrying passengers - running a red light is a threat to every car, driver, passenger and pedestrian in the vicinity. If neither the car nor its human driver can be relied upon to follow one of the simplest and most important rules of the road, it’s a pretty big problem for Uber’s test drives in densely populated cities. The car-sharing company had already begun trials in Pittsburgh earlier in 2016, where members of the public reported seeing one Uber driving the wrong way down a road, and another involved in a potential incident. There are no incidents on record with the city’s authorities, though.
The point of self-driving cars, of course, is to vastly improve road safety as well as efficiency. But they are in their infancy, and this incident is one of a handful to have been reported among those developing the technology. In February, a Google car hit a bus in California and was deemed to be at fault, and in May a Tesla Model S driver was killed while using the Autopilot beta feature, an incident that was blamed on speeding.
This article was originally published by WIRED UK