Tesla‘s Robotaxi fleet recorded no at-fault crashes with other vehicles between mid-June and mid-July, according to the latest incident reports published under the US National Highway Traffic Safety Administration’s (NHTSA) Standing General Order.
The pattern was first flagged by Tesla shareholder Sawyer Merritt on X this Monday.
Three incidents were logged, all dated June 2026 and all classified as property damage with no injuries reported. In each, the narrative Tesla submitted describes another vehicle striking a stationary Robotaxi.
The first happened in Austin.
A Robotaxi was stopped at a red left-turn arrow, in the dedicated left-turn lane of a four-lane street.
An SUV behind it signaled right, moved into the neighboring through-lane, and struck a car already traveling straight in that lane.
The SUV then continued into the stationary Robotaxi, hitting its right side. No one was riding in the Tesla.
The second happened in Dallas, when a Robotaxi was pulled over and in park on the right side of an alleyway. A vehicle parked to the right reversed and collided with the Robotaxi’s right side. The Tesla was again empty and stationary throughout.
The third, also in Austin, was more serious as one passenger was present inside the vehicle.
In both cases, the ADS made no maneuver that contributed to the collision, according to the narratives Tesla submitted to NHTSA.
Tesla said that “the construction truck stopped abruptly, forcing the Tesla ADS to stop in the area of reduced roadway width, then the truck reversed into the Tesla ADS’s front right.”
The third one, also in Austin, was more serious as one passenger was present inside the vehicle.
According to Tesla, its Model Y “was proceeding straight in a parking lot behind an SUV. As the SUV came to stop, the ADS came to a stop behind the SUV.”
Then, “the SUV proceeded to reverse and made contact with the ADS’s front.”
A Contradiction in the Third Filing
Tesla‘s own report on the Austin parking-lot crash is internally inconsistent. The narrative describes the Robotaxi as stopped behind the SUV when the SUV reversed into it. But the structured data field recording the vehicle’s pre-crash movement reads “Backing.”
A vehicle cannot simultaneously be stopped and reversing. Tesla has not explained the discrepancy, and it appears in the only one of the three June incidents with an occupant aboard.
Older Reports, Newly Unredacted
Several other reports published in the same batch carry a standard disclaimer: they don’t reflect a new incident, only the removal of confidential or personally identifying information from an older filing.
They are older narratives, previously redacted as confidential business information, now released in full — in one case alongside a correction to an erroneous airbag-deployment flag from a March filing.
Tesla was, for months, the only ADS operator redacting entire crash narratives from its NHTSA filings.
One such narrative describes a construction truck that “stopped abruptly, forcing the Tesla ADS to stop in the area of reduced roadway width, then the truck reversed into the Tesla ADS’s front right.” That incident predates the June reports.
Fleet Still Small Relative to Demand
The zero-at-fault stretch comes as Tesla works to demonstrate its safety record can support faster scaling.
Robotaxi now operates across seven US markets: Austin, Dallas, Houston, Miami, Orlando, Tampa, and the Bay Area.
Only the Bay Area fleet exceeds 500 vehicles, and it still requires a human safety driver under California regulations.
The unsupervised portion of the fleet remains far smaller, with Texas filings showing registered ride-hailing vehicles in the low hundreds.
That gap between market footprint and fleet size is also why Tesla is reportedly preparing a virtual booking queue for its Robotaxi app, designed to manage demand that already outstrips available vehicles.
The question of how to weigh Tesla‘s safety numbers against competitors also came up in TD Cowen’s decision to reiterate its Buy rating and $460 price target on Tesla last week.
Analyst Itay Michaeli’s note followed a session with autonomous vehicle specialist Alex Roy, who said it remains difficult to compare Tesla‘s safety levels to Waymo‘s because Tesla doesn’t disclose data on an apples-to-apples basis.
Roy also predicted that younger riders will eventually treat Tesla‘s safety level as “good enough,” even if it doesn’t match Waymo’s, once the service reaches sufficient scale.
The Bigger Picture
Tesla‘s active unsupervised fleet remains a fraction of Waymo‘s size, which logs a higher raw incident count simply by completing far more rides each week.
The company has been priotizing safety over deployment in scale.
“Our goals are very ambitious for Robotaxi, but we do need to be cautious about causing any accidents or causing any harm to anyone,” CEO Elon Musk repeated in the last earnings call.
The Chief Executive pointed to the asymmetric scrutiny autonomous vehicles face relative to human-driven crashes.
“Although there are, I think, 30,000 to 40,000 automotive deaths per year in the US alone, most of those do not generate any press,” he said. “If we injure even one person, it’ll be worldwide headline news, and regulators will immediately clamp down on our activities.”
Musk noted that “we’re going as fast as humanly possible in scaling Robotaxi, but while trying to ensure that we do not harm anyone at all, and ideally do not even run over a pet.”
