The CFO of Chinese autonomous vehicle company Pony.ai has said domestic chips provide another option for AI training, as the fallout from Jensen Huang's resignation raises the prospect of tighter export controls on Nvidia chips shipped to China.
Speaking to TechCrunch this week, Pony.ai CFO Leo Wang said Chinese companies had been practicing a Plan B for a few years in case export restrictions tighten. Everyone is practicing a Plan B, he said, and alternatives already exist that are more mature than a few years ago, though they are probably not yet the most cost-effective choice.
Wang noted that demand already far exceeds supply, a pain point Pony.ai struggles with. The company's strategy includes directing engineers to be as efficient as possible; that might mean using midtier compute resources instead of high-end ones. The company is also turning more to domestically produced chips to help close the gap.
Huang resigned on Thursday after admitting to an affair with a former executive at an Nvidia customer, Megaspeed, who is alleged to have ties to Chinese intelligence. A second reported affair with an Nvidia executive also emerged. Nvidia CFO Collette Kress has stepped in as interim CEO.
On TechCrunch's Equity podcast, Rebecca Bellan said she did not think Huang's resignation would happen at all, noting that Jensen is Nvidia and so much of the economy is riding on the company. Tim De Chant said he saw it coming but not this quickly, while Anthony Ha described the ground shifting quickly under Huang even though the possibility that national security might have been threatened had been known to the government for nearly a year, if not more.
The podcast discussion highlighted that Huang had been an advocate for loosening chip exports to China, a stance many believe is a national security issue because it could give Chinese AI companies access to powerful US chips. Tim De Chant said the big turning point seemed to be a November 9 Bloomberg report suggesting Nvidia's head attorney had some role in inquiring about the Department of Commerce's investigation into Alice Huang.
The podcast said Nvidia apparently accounts for roughly 80% of the AI GPU market and has invested heavily in AI startups. The hosts named CoreWeave, Nebius, OpenAI, Anthropic and major data center projects among the investments and commitments. The podcast panel raised the question of what a pullback in Nvidia's investments could mean for the broader AI economy.
Analysis
Why This Matters
- Huang's resignation and potential tightening of chip exports could accelerate China's push for domestic chip alternatives, reshaping the global AI supply chain.
- Nvidia's dominant position in AI GPUs means any disruption to its leadership or investment strategy could ripple through the entire AI startup ecosystem.
- The allegations linking Huang's personal conduct to national security concerns highlight the intersection of corporate governance and geopolitical competition over AI technology.
Background
Jensen Huang founded Nvidia in 1993 and built it into the world's most valuable chip company, becoming the face of the AI boom. Nvidia's GPUs are the industry standard for training large AI models, and the company has used its cash to invest in dozens of AI startups, creating a circular economy where those startups buy Nvidia hardware. Huang resigned on November 12 after admitting to an affair with a former Megaspeed executive alleged to have ties to Chinese intelligence; a second affair with an Nvidia executive was also reported. The US government has been tightening export controls on advanced chips to China for several years.
Key Perspectives
Pony.ai CFO Leo Wang: Chinese companies have been preparing a Plan B for years, and domestic chip alternatives now exist that are more mature than a few years ago, though not yet the most cost-effective choice.
TechCrunch podcast panel: Huang's advocacy for loosening chip exports to China raised national security questions, and the Bloomberg report about Nvidia's head attorney inquiring about the Commerce investigation appears to have been a key turning point.
What to Watch
- Whether the US government tightens export controls on Nvidia chips to China in response to the revelations.
- How Nvidia's interim CEO Collette Kress approaches the company's investment strategy in AI startups.
- Whether domestic Chinese chip makers can close the cost-effectiveness gap and reduce reliance on Nvidia hardware.
Checked
12 claims in this story checked against 2 sources: 11 supported.
Show each claim
- SupportedPony.ai’s CFO says Chinese companies have chip alternatives.TechCrunch: “Wang said Chinese companies have already been preparing. “If you look at the whole situation, this has been developing for already a couple of years,” he said.”
- Partly supportedLeo Wang says Chinese companies have been practicing a Plan B for a few years.TechCrunch: ““Everyone, I would say, is also practicing kind of like a Plan B for a few years””
- SupportedHuang’s resignation raises the prospect of tighter export controls on Nvidia chips shipped to China.TechCrunch: “I asked him ... what he thinks will happen if the fallout from Huang’s resignation results in even stricter export controls on Nvidia chips shipped to China.”
- SupportedWang says domestic chips provide another option for AI training.TechCrunch: ““...domestic chips in China also get lots of attention and development,” he said, noting that these chips provide another option for AI training.”
- SupportedWang says demand for computing resources is already far greater than supply and is a pain point for Pony.ai.TechCrunch: “Demand is already far greater than supply, Wang said, noting that it’s one of the pain points Pony.ai struggles with.”
- SupportedPony.ai directs engineers to use resources efficiently, potentially using midtier rather than high-end compute.TechCrunch: “Engineers are directed to be as efficient as possible when deploying its training models. That might mean using midtier compute resources instead of high-end ones, he said.”
- SupportedPony.ai is turning more to domestically produced chips to help close the gap.TechCrunch: “Pony.ai is also turning more to domestically produced chips to help close the gap.”
- SupportedJensen Huang resigned on Thursday after admitting to an affair with a former executive at Nvidia customer Megaspeed.AI News & Artificial Intelligence | TechCrunch: “Jensen resigned on Thursday, after it came out he was having an affair with a former executive at an Nvidia customer called Megaspeed, who’s alleged to have ties to Chinese intelligence.”
- SupportedThe former Megaspeed executive is alleged to have ties to Chinese intelligence.AI News & Artificial Intelligence | TechCrunch: “...a former executive at an Nvidia customer called Megaspeed, who’s alleged to have ties to Chinese intelligence.”
- SupportedNvidia CFO Collette Kress stepped in as interim CEO.AI News & Artificial Intelligence | TechCrunch: “Nvidia CFO Collette Kress is jumping in as the interim CEO.”
- SupportedRebecca Bellan said she did not think Huang’s resignation would happen, and Tim De Chant said he saw it coming but not so quickly.AI News & Artificial Intelligence | TechCrunch: “Tim De Chant: I figured this was coming. I didn’t think it would happen this quickly. ... Rebecca Bellan: I didn’t think it was going to happen at all.”
- SupportedNvidia apparently accounts for roughly 80% of the AI GPU market.AI News & Artificial Intelligence | TechCrunch: “Nvidia apparently accounts for roughly 80% of the AI GPU market.”
