Elon Musk slashes Tesla Robotaxi fleet goal from 500 to ~60 in Austin

Elon Musk announced last night that Tesla is planning to “roughly double” its Robotaxi fleet in Austin next month. While an expansion of the pilot sounds positive on the surface, a look at the actual numbers reveals that Tesla is missing its own “end of year” target by a massive margin.

Just last month, Musk explicitly stated that Tesla aimed to have 500 Robotaxis in Austin by the end of the year. Now, “doubling” the current estimated fleet suggests the actual number will be closer to 60.

We have been closely tracking the rollout of the “Tesla Robotaxi” pilot in Austin, which launched back in June using Model Y vehicles.

Unlike the “Cybercab” unveiled in October, these vehicles are standard Model Ys equipped with Hardware 4, and critically, they are not driverless. They are part of a “supervised” pilot, meaning a Tesla employee sits in the front passenger seat (or driver’s seat for highway stints) to monitor the system with a finger on a killswitch ready to stop the car..  

Advertisement – scroll for more content

The service has been plagued by availability issues. As we reported recently, users in Austin are frequently met with “High Service Demand” messages, with wait times often exceeding 40 minutes. It’s not necessarily because there’s really “high demand”, but because Tesla’s ‘Robotaxi fleet” remains tiny.

In response to complaints about the service being “essentially unusable” due to lack of supply, Elon Musk took to X (formerly Twitter) late Tuesday to promise relief:

“The Tesla Robotaxi fleet in Austin should roughly double next month.”  

For those frustrated by the wait times, more cars are certainly welcome. But for investors and analysts tracking Tesla’s autonomous driving promises, this announcement serves as a confirmation of a significant missed deadline.

How many Tesla Robotaxis are in Austin?

To understand why “doubling” is actually a disappointment, we have to look at what Musk promised just a few weeks ago.

During his appearance on the All-In Podcast, which aired on October 31, 2025, Musk was explicitly asked about the scale of the fleet. His answer was unambiguous:

“We’re scaling up the number of cars to… probably we’ll have a thousand cars or more in the Bay Area by the end of this year, probably 500 or more in the greater Austin area.”  

Let’s do the math.

Based on observations from the Austin community and tracking of the vehicle VINs and plate numbers, the current Tesla Robotaxi fleet in Austin is estimated to be around 30 vehicles. In fact, 29 different Robotaxi license plates were spotted in Austin.

If Tesla “roughly doubles” that fleet in December, they will have approximately 60 vehicles on the road.

That is a far cry from the 500 that Musk projected just weeks ago. In fact, it represents a shortfall of nearly 90% against the target.

This massive miss in deployment targets is particularly ironic given Musk’s recent comments about competitors. When Waymo announced earlier this month that it had reached 2,500 active robotaxis across the US (with about 200 in Austin alone), Musk scoffed, calling them “Rookie numbers.”  

Yet, the data shows that Waymo currently operates a fleet in Austin that is roughly 3x to 4x larger than what Tesla hopes to have after its expansion next month. And unlike Tesla’s pilot, Waymo’s Austin fleet is operating fully driverless, without human chaperones in the front seat.  

Electrek’s Take

Another clear case of Elon Musk’s shifting the goalposts in Tesla’s autonomous driving programs, something we’ve unfortunately become accustomed to with Tesla’s autonomy timelines.

Musk said “500 cars by end of year” just a few weeks ago. It shows he is just saying numbers and nothing is grounded in reality.

Let’s be real about what this means. It means the “unsupervised” dream is still stuck in “supervised” reality. Scaling a fleet to 500 cars when you need 1,000+ human employees to drive them (staffing multiple shifts) is an HR nightmare, not a software update. The fact that they are only getting to ~60 tells me that the “supervised” requirement is the hard limit on their growth right now.

With 7 crashes in the first few months of operations, with supervisors preventing an unknown number of additional crashes, I don’t see Tesla removing them anytime soon.

Let me rephrase that. I don’t see Tesla safely removing the supervisors anytime soon.

As this whole operation appears to be more about optics than safety, I can see removing them before it’s ready.