Tesla AI trainers said they do not trust the company's self-driving technology or the safety statistics it publishes to the public.
This internal distrust suggests a gap between the company's public promises and the actual performance of its Full Self-Driving (FSD) system. Because Tesla's $1.6 trillion market valuation [1] relies heavily on the promise of autonomy, these concerns from the employees who label and validate the data could impact investor confidence.
The employees, who work primarily at California facilities, said the company is far from safely delivering autonomous vehicles at scale [1]. They said the statistical methodology behind Tesla's safety claims is flawed, and that internal data suggests the company's promises are overstated [1].
These internal reports contrast with other data regarding the scale of the operation. Some reports indicate there are approximately four million FSD-enabled vehicles on the road [2]. This fleet reportedly drives between 30 million and 40 million miles daily [2].
While the sheer volume of data collection suggests a mature operation, the AI trainers said the resulting system remains unreliable [1]. The tension highlights a conflict between the quantity of data being processed and the quality of the safety outcomes produced by the AI training process.
Tesla's U.S. operations continue to push FSD as a core feature of its vehicle lineup. However, the trainers responsible for the system's accuracy said the safety stats are unreliable [1].
“Tesla AI trainers say they do not trust the company's self-driving technology”
The discrepancy between Tesla's massive data collection—millions of miles driven daily—and the skepticism of its own AI trainers suggests that more data does not automatically equal safer autonomy. If the internal staff responsible for validating the AI cannot trust the safety metrics, the company may face regulatory scrutiny or a valuation correction as the market shifts from valuing the potential of FSD to requiring proof of its safety.
