Spain's Traffic Regulator Logs 171,000 Miles of Tesla FSD With No Serious Incidents

A government registry from Spain's Dirección General de Tráfico shows Tesla's 30-car supervised test fleet has driven roughly 275,000 kilometers on public roads since November 2025, handing the company rare third-party evidence just as it pushes for European approval.

The technology was never the open question for Full Self-Driving in Europe. The binding constraint has been regulatory, and the most valuable thing Tesla could obtain was not another company-authored safety report but a clean record held by a government's own agency. Spain has now supplied exactly that, and the second-order consequence matters more than the headline number: once one European regulator carries a spotless 171,000-mile record on its books, that record becomes a reference template every other agency on the continent can lean on.

Key Takeaways

• Spain's Dirección General de Tráfico registry shows Tesla's supervised FSD fleet covered roughly 275,000 kilometers, about 171,000 miles, on public roads.
• Thirty vehicles logged those miles since November 2025 with no serious incidents recorded by the regulator.
• Operating under a national permit valid through November 2027, Tesla can run on any road in the country.
• Because the figures sit in a state regulator's own registry rather than a Tesla quarterly report, they function as rare third-party corroboration.
• Compared against rivals, Tesla dwarfs the field: bus operator ALSA logs about 1,400 kilometers and Level 2 tester Wayve runs just three vehicles.
• Every mile ran under Level 2 supervision, with a human driver fully responsible and expected to stay attentive throughout.
• Unresolved is the threshold Spain wants before shifting from supervised testing to customer-facing approval.
• Dutch authorities logged 1.8 million kilometers over 18 months, a rough marker for the scale European regulators may expect.

---

The Source Is the Story

The distance is impressive, but the provenance is what changes the calculus. Companies publish safety statistics constantly, and markets have learned to discount them because the party with the most to gain is also the party holding the pen. When the numbers instead live in a member state's official registry, the incentive structure inverts. The regulator has no reason to flatter the operator, and the data was collected under terms the regulator itself defined.

That distinction is why this data set carries weight that no quarterly report could. Tesla did not write these figures. It generated the behavior, and a government agency recorded the outcome. In a regulatory fight, that separation between actor and scorekeeper is worth more than any volume of self-reported miles.

Why Regulation, Not Engineering, Is the Wall

For two years the story of supervised autonomy in Europe has been framed as a technical race, but the real obstacle was always approval. The systems already function on public roads across other markets. What kept them largely blocked in Europe was a patchwork of national frameworks with no clean path from testing to deployment.

A clean government-held data set is the specific instrument that dissolves that obstacle. It converts an abstract argument about capability into a concrete record a regulator can cite. Once the question moves from "can it work" to "how much evidence do we need," the constraint has already begun to loosen.

The Reference Template Effect

Most coverage stops at the headline figure. The move worth watching is what a clean record does across borders. Regulators are institutionally cautious, and one of the cheapest ways to justify a cautious decision is to point at a peer agency that reached the same conclusion first.

Spain has now created that anchor. A Dutch, German, or French authority weighing its own approval can reference the Spanish registry rather than starting from zero. This is how a timeline everyone assumes is slow can compress: not through a single dramatic ruling, but through each agency leaning on the last until approval becomes the default rather than the exception.

Reading the Scale Honestly

The comparison inside Spain's own framework is stark. Tesla's fleet sits in the pre-deployment tier that permits more than ten vehicles, and its distance overwhelms everything else on the registry. ALSA's buses cover roughly 1,400 kilometers. Wayve operates three cars. Against that field, 171,000 miles is not incremental; it is the entire data set that matters.

Against a rigorous safety benchmark, though, the number is still young. A 30-car fleet over about seven months is modest next to the billions of miles that frame any serious statistical comparison with human drivers. Both things are true at once: dominant within the program, early against the standard that ultimately governs deployment.

The Fine Print on "No Serious Incidents"

The clean record deserves an asterisk that the headline tends to erase. Every one of those miles ran under Level 2 supervision, which means a human sat behind the wheel, fully responsible and expected to intervene. The system was assisting, not operating unattended, and that distinction is the whole game when translating a test record into a claim about autonomy.

There is a second gap. The traffic authority has not publicly defined what it counts as a serious incident. A clean record is only as meaningful as the threshold behind it, and until that definition is spelled out, the number invites more confidence than its terms strictly support.

The Car Sales Angle Nobody Prices

Robotaxi approval gets the attention, but there is a quieter payoff that accrues regardless of the regulatory outcome. Every supervised mile Tesla logs in Spain makes the underlying cars more capable and more valuable as products, whether or not driverless operation is ever cleared. The company is being paid in capability for work it would have to do anyway.

That framing matters because it removes the binary from the bet. If approval arrives, Tesla captures the robotaxi upside. If it stalls, Tesla still ships better cars into a market where driver-assistance quality increasingly drives purchase decisions. The downside case is not a loss; it is a slower win.

The Number Nobody Has Yet

The most important figure in this story does not exist on any registry, because Spain has not stated how many clean kilometers it wants before allowing Tesla to move from supervised testing to customer-facing approval. That undefined threshold is the real gate, and everything else is prologue to it.

The Dutch program offers a rough sense of magnitude: 1.8 million kilometers over 18 months. Whether Spain publishes a defined threshold, or another European regulator cites the Spanish record in its own deliberations, either signal would confirm the reference template is beginning to do its work. Those are the two developments worth tracking from here.