Nvidia founder and CEO Jensen Huang speaks during Nvidia Live at CES 2026 (Patrick T. Fallon/Getty Images)

Nvidia’s autonomous tech gives other automakers a chance to take on Tesla

Nvidia made a number of autonomous vehicle announcements at CES yesterday that should have Tesla worried.

1/6/26 9:00AM

Updated 1/6/26 1:54PM

Tesla is known for being an island. While most other carmakers and robotaxi services rely on an intricate web of hardware and software partnerships, Tesla does nearly everything in-house, from vehicle construction to the AI that powers its Full Self-Driving technology.

Enter Nvidia, whose autonomous driving software and hardware announcements yesterday underscore how Tesla’s once inaccessible approach to autonomy is becoming something other automakers can buy.

Presenting at the CES conference in Las Vegas yesterday, Nvidia CEO Jensen Huang unveiled Alpamayo, an open “reasoning” AI model family designed specifically for autonomous driving. The Alpamayo family of tools will introduce “chain-of-thought” models that can assess “novel or rare scenarios step by step.” In the press release accompanying the announcement, Huang said that the “ChatGPT moment for physical AI is here” and that he expects robotaxis to be among the first to benefit. Nvidia confirmed that Mercedes-Benz will begin shipping cars with the technology this year, offering advanced driver-assistance features comparable to Tesla’s FSD. (Mercedes’ former parent, Daimler, once held a 10% equity stake in Tesla.)

The AI chip designer also announced that it’s expanding its DRIVE Hyperion autonomous vehicle platform, bringing in a wide range of auto suppliers and sensor makers as it pushes the system as a ready-to-use foundation for self-driving cars and robotaxis. The platform bundles Nvidia’s latest Thor chips with cameras, radar, and lidar, centralized vehicle controls, and safety, simulation, and AI training tools — effectively packaging many of the pieces Tesla has built in-house into a system other automakers can use.

Taken together, the announcements show Nvidia turning autonomy into a platform business — one that could erode Tesla’s long-standing advantage by allowing rivals to buy the hardware, software, and safety infrastructure Tesla built for itself.

“I’m not losing any sleep about this,” Tesla CEO Elon Musk, who has had a notably chummy relationship with Huang, posted last night — one of several about the Nvidia announcements. “And I genuinely hope they succeed.”

I’m not losing any sleep about this.

And I genuinely hope they succeed.
— Elon Musk (@elonmusk) January 6, 2026

In another comment, Musk noted that there’s a years-long interval between when autonomous tech “sort of works to where it is much safer than a human,” suggesting that Tesla is way ahead of any would-be competitors. Indeed, Musk added, “This is maybe a competitive pressure on Tesla in 5 or 6 years, but probably longer.”

Tesla is down about more than 4% today, while Nvidia, up slightly earlier today, is flat. Aeva, a 4D lidar company Nvidia announced a partnership with, soared as high as 37%.

Claire Yubin Oh

isn’t it aironic?

10h

Stack Overflow’s forum is dead thanks to AI, but the company’s still kicking... thanks to AI

The platform is raking in millions of dollars in revenue, with AI an ironic new source of revenue.

Jon Keegan1/6/26

AI leaderboard maker LMArena hits $1.7 billion valuation

If you want to know who’s up and who’s down in the AI model world, look no further than LMArena’s leaderboard. The startup has just raised a $150 million series A fundraising round, with a valuation of $1.7 billion.

In seven months, LMArena has raised $250 million, according to TechCrunch.

The leaderboard started as a research project by cofounders Anastasios Angelopoulos and Wei-Lin Chiang when they were graduate students at UC Berkeley.

The public leaderboard — formerly known as “Chatbot Arena” — shows the results of human evaluations of AI models for various tasks. Users can rate which model did a better job on one task in a sort of blind taste test.

The leaderboard is a hotly contested proving grounds for new models, and the company occupies a powerful position in an industry that lacks independent, industry-standard evaluations.

LMArena lands $1.7B valuation four months after launching its product | TechCrunch

The leaderboard started as a research project by cofounders Anastasios Angelopoulos and Wei-Lin Chiang when they were graduate students at UC Berkeley.

The leaderboard is a hotly contested proving grounds for new models, and the company occupies a powerful position in an industry that lacks independent, industry-standard evaluations.

Rani Molla1/6/26

Uber jumps after unveiling Lucid robotaxi at CES

Uber shares jumped more than 5% after the company unveiled a production-intent robotaxi developed in partnership with Lucid and Nuro at the Consumer Electronics Show on Monday. The autonomous vehicle runs on Nvidia’s Drive AGX Thor computer. Nvidia itself announced a slate of autonomous hardware and software announcements at CES.

The companies said this fall that the San Francisco Bay Area will be the first market for the joint effort. The robotaxi is already being tested on public roads, with a commercial launch planned for later this year.

Uber + Lucid + Nvidia is just another example of the tangled web of partnerships in the autonomous driving space, where Nvidia is now becoming more and more prominent.

This is Uber's new robotaxi from Lucid and Nuro | TechCrunch

Uber + Lucid + Nvidia is just another example of the tangled web of partnerships in the autonomous driving space, where Nvidia is now becoming more and more prominent.

Rani Molla1/6/26

Meta delays international Ray-Ban Display expansion thanks to “unprecedented demand” and “extremely limited inventory”

Meta said today that it’s delaying the early 2026 international expansion of its Ray-Ban Display glasses because of “extremely limited inventory” and “unprecedented demand.” The company didn’t specify whether the issue was more supply or demand, but has previously insisted its smart glasses are a hit.

Waitlists for the smart glasses, which are controlled with a band you wear on your wrist, extend “well into 2026.”

“We’ll continue to focus on fulfilling orders in the US while we re-evaluate our approach to international availability,” the company wrote. Expansion had been planned for the UK, France, Italy, and Canada.

In order to buy the smart glasses, consumers must do an in-person product demo to ensure the tech is “properly fitted to you,” according to Meta. Demos in New York City are unavailable for the next few weeks, the company’s scheduling website shows. It also notes that “that due to high demand, the product may be sold out and unavailable for purchase after your demo.”

Jon Keegan1/5/26

Report: Returns on Nvidia’s Omniverse business are underwhelming

Today in Las Vegas, Nvidia’s Jensen Huang will be taking the stage at the CES conference to show off his company’s latest innovations. If you’ve watched any of Huang’s impressive jargon-filled keynote speeches, you’ll notice that he gets extra animated when talking about “digital twins.”

Nvidia’s “Omniverse” platform allows companies to use Nvidia software and hardware to run large-scale simulations of factories and assembly lines.

The idea is that you can train your robots in Nvidia’s simulated environment, saving huge amounts of time and money. Huang is making a huge bet on this nascent product, and according to a new report, it isn’t exactly taking off.

The Information reports that much to Huang’s anger and frustration, the Omniverse offerings have yet to generate significant revenue.

According to the report, despite a long list of customers who Nvidia says are using Omniverse, some say the software is hard to use and fails to allow for simulating robot interactions with complex objects. The report also says Nvidia shuttered its Omniverse Cloud service in August of last year due to a lack of demand.

Nvidia has certainly successfully blazed trails into new categories before, but considering Huang’s enthusiasm for the Omniverse offerings, it is a rare stumble for the chip juggernaut.

Nvidia’s Big Ambitions to Solve Manufacturing Shows Slow Returns So Far