‘Mass deployment is just around the corner’: Pony AI’s Peng on the robotaxi revolution

Since being listed on the Nasdaq last year, the autonomous driving startup Pony.ai has jumped to a valuation exceeding $7 billion — even as it still operates at a loss. James Peng, the Baidu and Google veteran who co-founded the company in 2016, aims to have more than 1,000 robotaxis on roads from Shenzhen to Dubai this year, and at least 10,000 within three years. And he has convinced well-established companies from Toyota to Uber to test his vehicle-agnostic hardware and software.

Peng is counting on differentiating Pony from well-funded rivals through low-cost technology, a long-term approach, and a culture that is quick to adapt. His ambition requires that Pony surmount challenges familiar to his competitors — overcoming technical hurdles; persuading users beyond early adopters to try unfamiliar technology; and convincing local leaders that self-driving cars are both safe and not unduly disruptive to their cities’ taxi drivers.

But Pony’s Chinese ownership also poses another challenge. In November 2023, a bipartisan group of US Congress members pressed it for details of any ties to China’s People’s Liberation Army.

Peng denies such ties, and says Pony’s Chinese expertise may give it advantages in the US — despite the hostile political climate.

This interview has been condensed and edited for clarity.

Andrew Edgecliffe-Johnson: You’ve passed some big milestones this year. Can you frame for us what this chapter of the story means for Pony AI?

James Peng: This is probably the most exciting moment for the company — not just for Pony, I think, but for the whole industry. Over a bit more than a decade, we’ve proven it’s safe and reliable to run vehicles fully driverless. And I consider this an inflection point where mass production and mass deployment is just around the corner. We’ve achieved one, and now’s the time to add zeroes behind it, so that we can be thousands, tens of thousands, and all that. This is the most important: that we make sure that we have a [number] one in front of the big number, and then from now on, everything is all about scaling up.

Waymo has one of the world’s largest companies behind it in Alphabet, and Zoox is backed by Amazon. What makes you think that Pony can compete with such deep-pocketed and technologically advanced groups?

It’s through fast execution, and because mobility is such a huge industry. If this was a low-entry-barrier industry, we’d probably [find it] very hard, but because this is such a high-ceiling industry, and it’s so big, it has the capacity for many companies to survive. So I think this gives us the breathing room, and we have been able to get more and more resources as we prove ourselves to beef us up and be able to compete and face those giants.

What has been most important in working with governments to gain access to the cities that Pony wants to operate in?

We treat the governments as long-term partners, so a lot of the collaboration is an iterative process. Another thing is that we don’t put our eggs in one basket. We work with several cities, and we keep long-term relationships with them. At the same time, we also sort of use them to compete with each other. They want to be innovative, they want to solve some of the mobility challenges in their cities. But they view this as a disruption to existing mobility systems and also, potentially, long-term, to job security for the local labor force. So it’s always the balancing game between the two. And in those cases we can always use the success in one city to set and push for the other city.

Do you agree that this is disruptive to the labor force, that this is the end of taxi drivers’ jobs?

It’s not like this is the first time this has happened in this industry. It happens in many industries — it used to be that all the phone lines were connected by operators. But I think the good thing is that the transition definitely will not happen overnight. And we are creating new jobs, new types of jobs — the maintenance guys, the safety operators, and all that. So I think, as a society, as long as it’s long-term, there’s always a way to create new jobs and absorb some of the [disrupted] labor force.

There are also national politics at play. In May, the heads of two congressional panels called on the SEC to delist a number of Chinese companies, including Pony. They were raising national security concerns, saying that you have ties to China’s military. Were their facts right?

No, we’re not directly working with any military-related entities. I think some of our partners might have a connection to the military, but we never have, ever, [had] any direct collaboration. Some of our partners probably have different arms [that] work with the military, but they’re not autonomous-related. So I think those facts were definitely not right.

How are you trying to change that conversation in Washington?

Our voice is not that big in Washington, so there’s not too much we can do. What we can do is… minimize the risks. As an example, we’re definitely not going to work directly with the [Chinese] military. We are not doing anything that might potentially interfere with certain regulations. And, in terms of the tariffs, we are trying to minimize and have less trade between China and the US. We’re definitely not transferring data between the countries. So, essentially, we’re trying to do as much as we can to minimize the risks.

Do the politics of US-China relations make the US a harder country for Pony to succeed in?

Our current policy is that we are maintaining our R&D efforts in the US, but we are minimizing our commercialization efforts in the US. I’ve been in the US for almost 30 years and I think the policy is always up and down, so maybe when it changes, we’ll get in again.

[Our] immediate focus is in China, because the policy [there] is probably the most advanced. We’re definitely looking at Korea, Japan, the Middle East, Europe, potentially Australia. I think outside of this, probably, it’s going to be hard.

So, is it possible that the US falls behind China in this industry?

No, I think China and the US will go different ways. They have different approaches. The US, technology-wise, is definitely quite advanced. In China, the traffic challenge is higher, it’s more crowded. So that forces us to be more advanced in dealing with [risks that are considered long-tail] in the US.

The problem is more chaotic, so our technology has to be adaptive to the roads in China. That probably gives us an edge. In South Korea, I think we definitely have a technical edge compared with our US peers, because the road conditions there are more similar to what’s in China than the US.