Self-driving cars are no longer confined to the realm of science fiction, thanks to companies like Waymo rolling out commercial robo-taxi services in multiple cities. And many automakers, large and small, are looking to bring the same level of self-driving capabilities to privately owned vehicles over the next few years.
If this happens, what can personal AV do? Maybe you can send them to buy food or run errands — that’s what the company Autolane hopes to do. Or, if you’re Tesla CEO Elon Musk, owners might want to deploy their cars around the world, offering rides to others and creating a tidy, almost effortless side hustle.
Hebron Sher, co-founder of Dallas-based Zevo, wants to take advantage of all of the above. The company has been operating an EV-only car-share fleet for a little over a year, and is now adding robotaxis, starting with a new company called Tensor.
Tensors are a bit of an odd duck. The startup, which was spun out earlier this year from a previous Silicon Valley startup called AutoX that also had operations in China, boldly claims to be selling fully self-driving cars to consumers for the first time in 2026. Zevo says it will purchase up to 100 Tensor cars and add them to its network.
Few details have been released about timing, and Tensor still needs to show it can build cars at any scale and reliability, a major hurdle that has stumbled on many other new entrants in recent years. (The company said in a statement that “the path to shipment is driven by the pace of global regulatory approvals, not product limitations.”) If it goes well, Zevo customers will be able to rent Tensor AVs, essentially amounting to a kind of decentralized robotaxi service.
The announcement is similar to the many whirlwind promises made a decade ago when self-driving car hype first peaked. But now that there are actual robotaxis on the road, the idea seems a little more concrete this time around.
“Tensor’s vision is to build a future where everyone owns their own artificial general intelligence, a personal AGI that enables more time, freedom, and autonomy,” Hugo Fozzati, Tensor’s chief business officer, said in a statement. “For us, this partnership with Zevo will not only allow us to sell our vehicles in bulk, but also allow individuals and micro-entrepreneurs to participate and profit from the AV business in this AI era.”
tech crunch event
san francisco
|
October 13-15, 2026
This isn’t the first time Mr. Shah has bet on an unproven company. In October, Zevo announced a non-binding order for 1,000 vans from permanently endangered EV startup Faraday Future. After struggling for a decade to make and sell its flagship FF91 luxury SUV, Faraday recently resorted to importing a more affordable electric van from China and assembling it in the United States. The company plans to assemble its first electric van later this month.
Mr. Shah is willing to bet on these companies for two reasons. One is cost. Both Tensor and Faraday Future have non-existent and bad track records, respectively, and would like to get an attractive deal. More importantly, these companies are more willing than large OEMs to allow Zevo to integrate tightly with the vehicle’s software, Shah said. This is an important factor, especially when allowing peer-to-peer sharing of robotaxis.
Not only is it difficult to give traditional automakers similar flexibility, but their “technical weirdness is the worst,” Shah said. “I think they really enjoyed the experience of the synergy of startups and startups, kind of mixing and mingling and saying, ‘Let’s build solutions together…'” he said of the newcomers.
Shah paints the deal with Tensor as a win-win for both companies. Tensor has a chance to win early customers and prove that its robot car is real and can do what it promises. Zevo wins great deals because of its tight software integration.
But he admits it’s a “calculated risk”.
“I think the fact that we encourage startups to take risks is what makes America great,” he said.
This article has been updated to reflect that AutoX’s headquarters are in Silicon Valley.
Source link
