Tesla's Cybercab launched on a certification NHTSA never approved
Every automaker in America vouches for its own vehicles. Tesla just found out what happens when the vouching gets creative — and a regulator that likes autonomous vehicles anyway starts reading the fine print.
The probe is about paperwork, not driving
When Tesla put its first Cybercabs on Austin's streets on September 3, it did so without a federal exemption, without a waiver, and without steering wheel, pedals, or sideview mirrors — three features the Federal Motor Vehicle Safety Standards require on ordinary cars. Instead, Tesla told NHTSA it had certified the two-seater as compliant with all applicable standards, relying on determinations that certain standards simply don't apply to a vehicle designed never to be driven by a human. Within hours, the agency opened an Audit Query — docket AQ26002 — to examine "the process and technical data on which Tesla relied when certifying the Cybercab and related issues," including the extent to which that certification rested on deciding which rules count. The query covers an estimated 1,000 vehicles, of which 45 are registered with Texas authorities. Tesla has kept operating the Austin fleet, and an Audit Query is not a recall; it is an examination of whether a company's homework was real.
The distinction matters more than the viral videos this week. Riders have been filming Cybercabs stopping in live traffic lanes instead of at the curb — an embarrassing routing failure we covered this morning. But a mid-road stop is a software complaint. What NHTSA is asking about is a legal one: whether a company can take the position, on its own authority, that federal safety standards written for human-driven cars are inapplicable to its driverless one, then sell rides on that interpretation. No exemption was sought. The certification was the interpretation.
The loophole Tesla is testing
Self-certification is the quiet foundation of the entire US auto market. NHTSA writes the performance standards; automakers certify their own compliance; the agency polices the system after the fact through investigations and recalls. The regime works because the interpretations involved are usually boring — a crash-test measurement, a lighting specification. The Cybercab stress-tests it at the edge: the interpretation in question is not about how well a component performs but about whether the component must exist.
There is a lawful path for vehicles that omit required equipment, and Tesla did not take it. It runs through NHTSA's exemption process, which caps volumes and takes years — Zoox spent the better part of a decade on that road and only received final federal approval to charge riders this July, a sequence we followed as Waymo opened new cities while Zoox tested in Houston. Tesla's move skips the queue entirely: deploy now, argue later, and let the political environment do the waiting for it. The bet is not insane. The Transportation Department has proposed removing the manual-controls requirements for autonomous vehicles outright, and NHTSA Administrator Jonathan Morrison framed his own agency's probe in Innovation Frontier language, saying the agency "fully supports the safe development and deployment of automated vehicles" while insisting laws be followed. The regulator that wants to legalize exactly this design is now investigating whether Tesla jumped ahead of the legalization.
That is the uncomfortable part for Tesla — and the reason the probe has teeth. An Audit Query opened by a sympathetic administration means the industry-friendly path (new rules) and the enforcement path (this query) are now racing. If the rule rewrite lands first, the certification question evaporates. If the query escalates first, Tesla's certification becomes the test case for how much interpretive latitude self-certification actually gives a manufacturer — and the precedent lands on every robotaxi company behind it.
Who wins the slow way
The companies Tesla outflanked come out looking prudent. Zoox and Waymo cleared or worked within the existing machinery — exemption petitions, state permits, staged rollouts — and absorbed years of delay for it. Waymo keeps opening cities on the conventional path; Uber and Wayve just beat Waymo to commercial service in London under a regulatory regime that is stricter still. The Cybercab strategy implies all of that caution was a tax paid by suckers. If it works, every operator will demand the same interpretive latitude, and the exemption process becomes theater. If it fails, the slow way gets vindicated and NHTSA's process becomes the only game in town again.
There is a third group losing quietly: the consumers Tesla is courting as fleet owners. The company has been asking the public whether they want to buy and operate Cybercab fleets of their own — inviting individuals to become robotaxi entrepreneurs while the vehicle's certification basis sits under federal examination. A fleet buyer is making a capital bet on a legal theory. That risk is not disclosed on any order page.
The skeptic's case
The obvious defense of Tesla deserves a fair hearing: the FMVSS were written for cars with humans at the controls, and a vehicle with no steering column arguably cannot "comply" with requirements that presuppose one. Regulators have known this gap exists for years; the proposed rule changes are an admission of it. From that angle, Tesla is not gaming the system — it is forcing the system to confront rules that no longer map onto the technology, the way every earlier certification fight eventually did.
But the skeptic's counter is about process, not substance. The mechanism for updating rules runs through notice, comment, and agency determination — precisely because letting each manufacturer decide which laws apply to it substitutes a company's commercial timeline for the public's safety review. Tesla's own deployment data makes the point awkwardly: the first week produced vehicles stalling in traffic lanes, the kind of operational evidence that suggests the "simple" parts of the problem were not as solved as the certification assumed. A company that wants regulators to trust its judgment on which standards are inapplicable does not help its case by stopping in the middle of the road.
What to watch: whether the Audit Query escalates into a defect investigation or a certification denial, whether the proposed AV rule rewrite lands and moots the question, and whether Tesla slows fleet-participation sales while the paperwork fight runs.
If the rules are outdated, who should get to decide they don't apply — an agency writing new ones, or the company selling the car? Tell us in the comments.
Sources: NHTSA · TechCrunch · The Verge · The Verge — launch report · Tesla fleet-participation ask (TechCrunch)
