Google’s self-driving car software is seen as a driver

GOOGLE’S artificial-intelligence system will be interpreted as a driver by federal regulators, a step towards compliance that would help the tech giant’s self-driving cars hit US roads.

The National Highway Traffic Safety Administration (NHTSA) agrees with Google that its cars “will not have a ‘driver’ in the traditional sense that vehicles have had drivers during the last more than 100 years”, the agency said in a letter last week to Chris Urmson, the director of the company’s self-driving car project.

Google asked NHTSA in a November letter for interpretation of safety standards in cars it seeks to produce without traditional controls, such as a steering wheel or throttle and brake pedals. “If no human occupant of the vehicle can actually drive the vehicle, it is more reasonable to identify the ‘driver’ as whatever is doing the driving,” Paul Hemmersbaugh, the NHTSA’s chief counsel, said in the letter.

“The government will have to study the capabilities of Google’s autonomous tech and be convinced it doesn’t represent a danger,” Karl Brauer, an analyst with researcher Kelley Blue Book, said. “That might take a while if it involves a permanent change to the vehicle code, but the process of getting an exemption would be easier and quicker.”

Alphabet’s Google is getting mixed signals from federal and state regulators on the path to putting its self-driving cars on the road.