• Tesla, for instance, claims it will have a million self-driving robotaxis by next year.
• "We want to be safe. And so that's what's motivating us," Barra tells "Axios on HBO." "We understand this is life-changing technology."

What's happening: GM's self-driving unit, Cruise, set a 2019 target to launch robotaxis "at scale" in San Francisco, but that timetable already seems in jeopardy.

• Test vehicles have faced various technical glitches, according to one report in The Information (subscription required), including an embarrassing failed demonstration for the CEO of Honda, a Cruise investor, in April.
• Though not unusual for technology still in development, the reported problems underscore the challenge of deploying self-driving cars following early hype-fueled advancements.
• GM's own message has changed over the past year, from telling investors it is "on track" for commercial launch in 2019 to "we'll be gated by safety." (Company officials say the timeline hasn't changed.)

On "Axios on HBO," CEO Mary Barra says the most important thing is making sure the technology is ready to drive on its own, without a backup safety driver.

• Waymo, the Google spinoff that is Cruise's biggest competitor, launched a commercial ride-hailing service near Phoenix last December called Waymo One. But riders who summon a Waymo van still have a backup safety driver in the front seat.
• "If you can demonstrate that your technology is safer than a human driver, that's what we think then makes it acceptable to put out on the road," Barra said.

Even without a safety driver, GM's first AVs for public use will likely be equipped with a steering wheel and pedals because the U.S. Department of Transportation hasn't acted on its request for an exemption of motor vehicle safety laws that require manual controls.

While not committing to a launch date, Barra says she doesn't regret the 2019 target: "It's a rallying cry. And I think it's been motivational."

What we're watching: While Cruise continues to improve the technical capabilities of its cars and weighs the timing of their deployment, GM could draw lessons from the problems faced by scooter companies in San Francisco.

• Companies like Bird and Lime were booted off the streets and only allowed to return by applying for highly regulated permits from the San Francisco Municipal Transportation Agency.

The bottom line: While weighing whether its self-driving technology is ready, GM must also consider whether consumers are ready to ride in a car when no one's behind the wheel.

What to watch: VW is likely to take a stake in Argo AI, the Ford-backed self-driving technology startup.

Go deeper: Ford reshuffles at the top

Between the lines: Speaking Tuesday to an audience at the Economic Club of Washington, D.C., Uber CEO Dara Khosrowshahi said...

• "We really want to move from being a ride-hailing app to essentially being your transportation partner."
• His vision is to have a multi-modal Uber transportation network that, with a push of a button, helps people plan how to get from A to B, balancing tradeoffs like time, convenience and price.
• Uber doesn't just want to move people — this morning, the company announced it will start testing drone food delivery in urban areas.

Yes, but: Serious obstacles stand in the way of flying cars, including regulations, infrastructure and air traffic management issues, not to mention consumer acceptance and safety, Deloitte's Robin Lineberger tells Axios.

• "They will be a legitimate part of a multi-modal transportation system, in 20 to 30 years," he says.

Driving the news: At its third annual Uber Elevate conference, the company showcased vertical takeoff and landing models from 5 potential suppliers, including a full-size model from Bell that looks like a cross between a plane and a helicopter.

• Attendees could climb into a mockup of a cabin interior built by the French aerospace company Safran or strap on virtual reality headsets to experience what a flying taxi ride would be like.
• Uber said Melbourne, Australia, would be its first international pilot site for Uber Air, after Los Angeles and Dallas-Fort Worth in the United States, starting in 2020.
• And it showcased 16 potential designs for urban skyports, where flying taxis would take off and land.

Uber wants to "get the industry moving and designing these vehicles so that they can be available for urban transportation," Khosrowshahi said. "We want the pricing of this service to be ultimately available for the masses versus just the elites."

So far, it looks out of reach for most. A new 8-minute Uber helicopter shuttle between Manhattan and John F. Kennedy Airport is expected to cost $200 when it launches next month.

• In the wake of a helicopter crash this week on the roof of a building in Manhattan, some lawmakers want to see a ban on such flights.
• And some argue it overshadows existing transit reform efforts.

The bottom line: The market for air taxis is expected to grow from $3.4 billion in 2025 to $17.7 billion by 2040, according to Deloitte, and for newly public Uber, which lost $1 billion in the first quarter of 2019, that opportunity is hard to pass up.

• "We think it’s time to lean forward," Khosrowshahi says. "The business is well positioned to profit. But the next 2, 3, 4 years are going to be about growth."

Go deeper:

• A sky full of driverless flying cars in just a decade
• High hurdles for flying cars

Editor's note: This piece has been updated to clarify the timing of Uber’s $1 billion loss.