China’s self-driving truck leaders say AI success won’t speed up rollout – here it is

BEIJING — While AI updates hit the headlines every few weeks, those advances aren’t enough to get self-driving cars on the road any time soon.

This is according to China’s private trucking companies, which say that the development of major language models, from Anthropic’s Claude to China’s DeepSeek, has little impact on the timeline of vehicle shipments.

“The best languages ​​in the world [expert] it doesn’t mean you’re a good driver,” Pony.ai CEO James Peng told reporters last week. “AI is a very broad term. They are completely different things. Of course … there is no consistency.”

“When we process language, we play games, when we drive we use different skills,” he said.

Automated driving uses artificial intelligence to mimic a human driver through a combination of sensors, chips and algorithms. But the real-world training data needed is very different from what powers big language models like OpenAI’s ChatGPT, which require so-called global models.

Inceptio, the self-driving truck startup, is sticking to its timeline for a commercial milestone of mid-2028, which could be disrupted by broader advances in AI, CEO Julian Ma told CNBC.

By the third or fourth quarter of 2028, he expects Inceptio to have collected 5 billion kilometers (3.1 billion miles) of truck driving data in China — enough to allow full-duty trucks to hit public roads.

With 5 billion kilometers of driving data collected, AI can extrapolate that from 50 billion kilometers into a global model — enough to allow a heavy-duty truck to drive itself, Ma said. He expects the trucks to start operating without people inside certain areas of the country.

Achieving that goal in about two years is already too soon, he said, noting that for driverless trucks to become a widespread reality, they will need partnerships with manufacturers and regulatory approvals — in addition to technology.

Autonomous vehicles rely heavily on data about driving on roads. Like robotics companies, self-driving truck operators run human-powered tests to collect data for safety training.

Inceptio has recorded the most autonomous truck miles in the industry, surpassing US rivals, according to ARK Invest’s Big Ideas 2026 report in January. During that time, the company had driven 250 million miles – more than any private Chinese driving company. Pony.aiholding second place with 4.2 million miles.

US-based rivals Aurora, Kodiak and Gatik rounded out the top five, with 8.9 million miles, according to the report.

Inceptio’s Ma said that by the end of April, the company’s trucks had traveled 700 million kilometers (434.96 million miles), and they are targeting 1 billion kilometers (621.4 million miles) by the end of the year. He said the company could use AI to identify which conditions to focus on in order to collect test data.

At the Beijing Auto Show, Pony.ai also announced an upgrade to its PonyWorld 2.0 AI model to improve its ability to collect specific data and train the model more effectively. The company, which already uses robotaxis in China and other countries, has introduced a fully driverless truck that produces large batteries. CATL.

While China has 5-year development plans that place more emphasis on technological goals, Ma said that it is often companies that lead the way in driving innovation.

“We make it happen,” he said, before regulators see the technology in action and are confident enough to provide policy support.

But it is clear that there is still a long way to go before you see trucks and cars running across the country without drivers.

“Cars are actually a very challenging area for AI, and it exceeds the difficulty of integrated AI to some extent, because it involves safety,” Ma said. Featured AI includes humanoid robots.