On June 19, a terrible tragedy shook a peaceful residential neighborhood in Katy, Texas. An electric sedan suddenly left the roadway at very high speed before crashing into a house, instantly killing its occupant, Martha Avila Mantilla, aged 76. Very quickly, the affair took on a national scale, because two versions clash : that of the driver, who claims that the assisted driving system is responsible, and that of Tesla, who accuses flagrant human error.
What really happened during the accident?
According to initial information, the driver, Michael Butler, immediately told investigators that he had used autopilot activated at the time of the facts. Law enforcement quickly ruled out alcohol and described the driver as fully cooperative. A camera video surveillance a neighbor captured the scene, showing the car hurtling down the street without any sign of braking before the extremely violent impact.
Yes, this makes no sense. FSD drives slowly through neighborhood streets and this was a high speed crash!
— Elon Musk (@elonmusk) June 22, 2026
Faced with these accusations, Tesla opposed a formal denialbased on telemetric data from the Tesla Model 3. Ashok Elluswamy, the director of AI at Tesla, publicly stated that the driver had “ canceled manually automatic piloting with accelerator pedal depressed 100% », reaching a speed of 117 km/h. Elon Musk himself called the hypothesis of FSD failure absurd, arguing that the system is programmed to drive slowly in residential areas.
What is the scope of the official investigation?
To decide between these two narratives, the US National Highway Traffic Safety Administration opened a thorough investigation. This “special crash investigation” carried out by the NHTSA (National Highway Traffic Safety Administration) is the agency’s most detailed procedure, often triggered when emerging technologies are involved. Investigators will analyze the vehicle’s black boxes, which alone will be able to confirm or deny the assertions of both parties.
This drama takes place in a context where Tesla’s technology is closely scrutinized. There system reliability Autopilot and FSD (Full Self-Driving) are the subject of much criticism and a broader investigation by the NHTSA. By 2023, more than two million vehicles had been recalled for a software fix to improve driver attention monitoring. American senators also questioned the agency, denouncing Tesla’s assertions on safety of its drivers based, according to them, on incomplete or misleading data.
Yup. In this case, the driver manually overrode self-driving by pressing the accelerator all the way to 100% of the accel pedal in this residential area. They reached a speed of 73 mph during the crash, and had the accelerator pressed even after the crash.
— Ashok Elluswamy (@aelluswamy) June 22, 2026
How does Tesla usually handle these crises?
The defense strategy of Tesla relies almost systematically on its unique ability to remotely access vehicle data. The manufacturer can analyze in detail the speed, the angle of the steering wheel, the pressure on the pedals and the state of the driver assistance systems a few seconds before an impact. This technological transparency allows him to quickly counter accusations that he considers unfounded.
This is not the first time that the company has used this advantage. A notable precedent took place in 2022, when a Tesla employee died at the wheel of his own Model 3. His family had blamed Autopilot, but the data, corroborated by the police, showed that the driver was drunk and that the system had been deactivated well before this fatal accident. The conclusions of the NHTSA investigation into the Texas tragedy are therefore eagerly awaited to establish responsibilities.
