Even though startups have raised tens of billions of dollars based on promises to develop cars that can drive themselves, some industry executives and experts think that human supervisors may be permanently needed to help robot drivers in trouble.
Reuters correspondent Nick Carey.
"The problem with truly autonomous vehicles is that robot drivers are just not as quick as human drivers in assessing situations and making decisions as they drive. It's very difficult for machines to view the world that we do and assess, see what's happening, see what's coming and make decisions."
Many AV startups today use humans as remote supervisors, alongside safety drivers sitting behind the wheel.
Human supervisors could be sitting tens of hundreds of miles away monitoring video feeds from multiple driverless vehicles all at once.
Those remote humans are an additional expense, but they can help self-driving cars handle edge cases.
These could include something as basic as an unfamiliar set of lane closures during road construction or erratic, unpredictable behavior by pedestrians or other drivers.
Autonomous vehicle company, Cruise, says it relies on human supervisors less than 1% of the time.
But CEO Kyle Vogt told Reuters that their presence provides invaluable peace of mind for drivers.
"The CEO of Cruise says that human beings may be needed forever in their vehicles in the short term and in the medium term. Those humans need to step in. About 1% of the time to tell the autonomous vehicles what to do in the long term. They may be needed in fewer cases, but will certainly be needed to help the vehicles, but may also be needed to help give human passengers peace of mind and also help them in the event that they end up in trouble.''
This was the first time the self-driving unit of General Motors acknowledged the long-term need for remote human operators.
In June 2022, a Cruise vehicle was involved in a crash in San Francisco that left two people injured.
GM recalled and updated software in its self-driving vehicles and said the unusual scenario would not recur after the update.
For some, the idea that human supervisors could be here to stay raises more doubts about the technology.
Truly autonomous vehicles are far behind the optimistic rollout schedules predicted just a few years ago.
And the undelivered promise of true autonomy has raised the stakes for the AV industry.
"We're now several years beyond where the likes of Tesla or Cruise are promising that they'd have driverless robo taxis operating in large numbers on the roads. Investors have put serious money into this, and these autonomous vehicle companies have yet to deliver. So the problem for now is that it's difficult, harder and harder to raise more money for people who've already paid an awful lot to make this happen. And this is going to cause problems for some of the autonomous vehicle companies. Those with deep pockets or with big backers are likely to continue. But there's an open question as to the survival of some of the smaller players in the long term if they can't raise more money."
Overall investment in future mobility startups has slowed, according to investor website PitchBook.
Mike Wagner, CEO of Edge Case Research, which helps autonomous vehicle companies assess, manage and insure risk, warned that if these companies don't succeed over the next two years, they'll cease to exist.
So, are autonomous vehicle companies going to have to think of other routes to take?
"The next thing for autonomous vehicle companies is that they have to start finding ways to make money or make a revenue generating service, doing something like delivering goods or operating in mines or operating in remote areas where at slow speeds, where their vehicles don't encounter many problems or many pedestrians, anything that's going to bring in money and hopefully generate a profit is going to be crucial because those autonomous vehicle companies need to raise more money. And if they're not making any kind of profit or generating revenue, that's got to be very difficult."
