Earlier this month, the ride-hailing service Uber rolled out its inaugural fleet of self-driving cars in Pittsburgh. But as sci-fi as the concept sounded -- a fully automated car picking you up curbside to drive you wherever you want to go -- Uber remained cautious about the program's experimental nature.
For starters, the cars aren't completely driverless: Uber drivers still sit behind the wheel in case of emergencies. In addition, Uber passengers had to sign a liability form in case they were injured or killed in an accident while riding in these self-driving cars, according to documents obtained by The Guardian.
"I acknowledge that some or all of the [autonomous vehicles] in which I ride are in a development phase," read one of these waivers. The document, from June 2016, also stated that these cars "are experimental in nature, and that riding in an AV may involve the potential for death, serious injury, and/or property loss."
These days, Uber does not require the waiver. The self-driving cars are insured up to $5 million per incident, Uber said in a statement.
But the waiver is telling of how nascent the program is, and what precautions riders should consider before entering a self-driving car.
Though automated cars are seen as a relatively safe inevitability, the jury is still out on when they will be safe enough to be ubiquitous and deployed to mainstream audiences.
In February, a Google-owned self-driving car hit the side of a bus, in what Google called "tricky" traffic conditions. In June, a Tesla Model S, which had its autopilot engaged, was involved in an accident that resulted in the fatality of the driver.
Despite these incidents, Uber's self-driving program doesn't look like it will slow down. One of its cars was recently spotted in San Francisco and the Advanced Technologies Center (Uber's research and development arm for self-driving cars) states that it plans on making the "process of moving real assets and real people safer and more efficient now, and in the decades to come."