The Tesla autonomy debate has always been part engineering argument and part social media war, so when a clip that changed the tone appeared on X, I wasn’t surprised.
Nvidia CEO Jensen Huang said he believes “Tesla’s stack is the most advanced autonomous vehicle stack in the world” and that he is “pretty sure they were already using end-to-end AI,” as quoted at CES 2026 in his interview with Bloomberg.
Sawyer Merritt, a widely followed Tesla investor, posted the clip with the quote on X (formerly Twitter) and credited Bloomberg’s Ed Ludlow for the interview.
That sentence accomplished what years of Tesla blog posts and earnings statements have struggled to achieve. He took the core of Musk’s autonomy speech and put it in the mouth of the largest AI hardware provider on the planet.
Merritt’s post captured Huang’s full thinking, including his point that whether Tesla’s AI did explicit “reasoning” was “secondary” to the fact that it was already end-to-end. That’s exactly how Tesla has described its full self-driving stack since its rewrite in 2023.
I’ve been writing about Tesla long enough to know how quickly a quote like that becomes canon.
Within hours, other accounts cut the clip, added captions like “Nvidia CEO admits Tesla’s leadership in FSD is phenomenal,” and treated the phrase “world’s most advanced stack” as the last word in discussions about whether Tesla is behind Waymo or other robotaxi projects.
StockSavvyShay neatly summed up the mood, posting on X that Huang had praised Tesla’s autonomy stack and said the company was “doing a great job.”
The second wave of posts that caught my attention came from Wolf Financial, which often acts as an amplifier for clips on market movements. “Jensen Huang, CEO of $NVDA, was just asked about the difference between Nvidia’s Alpamayo and $TSLA FSD. Here’s his full answer,” Wolf wrote, linking to the longer segment where Huang describes Nvidia’s platform.
In that exchange, Huang breaks autonomy down into four layers: data centers and networks, AI infrastructure, the driving stack itself, and the car hardware that runs it. Wolf and other commenters seized on the part where Huang acknowledges that Tesla addressed a “real end-to-end problem” and built an entire stack on Nvidia chips, something he called “remarkable,” given Tesla’s relatively small team.
Those details resurfaced in YouTube breakdowns with titles like “Nvidia CEO Reveals Tesla’s Biggest Threat Yet,” which argued that Alpamayo validates Tesla’s camera- and neural-network-first approach, even as it opens the door to competitors.