Tesla submitted self generated safety data to regulators in Sweden and the Netherlands which independent traffic safety experts say is built on flawed methodology and could constitute misleading advertising. Tesla's own data labelers—the workers who train the AI system—do not trust the technology enough to use it th...

Create a landscape editorial hero image for this Studio Global article: What did a Reuters investigation reveal about Tesla's self-reported Full Self-Driving (FSD) safety data submitted to European regulators, ho. Article summary: Here is the full picture based on Reuters' reporting and independent sources.. Topic tags: general, news, general web, user generated. Reference image context from search candidates: Reference image 1: visual subject "# Tesla’s own AI trainers don’t trust ‘Full Self-Driving’ or its safety stats, Reuters finds. A major Reuters investigation published today reveals that Tesla’s widely touted “Full" source context "Tesla's own AI trainers don't trust 'Full Self-Driving' or its safety stats, Reuters finds | Electrek" Reference image 2: visual subject "# Tesla’s own AI trainers don’t trust ‘Full Self-Driving’ or its safety stats, Reuters finds. A major Reuters inves
Tesla’s campaign to bring its Full Self-Driving (Supervised) system to European roads is colliding with a deepening credibility problem. A Reuters investigation published on June 15, 2026, found that the company submitted self-generated safety data to regulators in Sweden and the Netherlands that independent traffic-safety experts believe could amount to misleading advertising . The report, built on months of reporting and internal correspondence, lays bare a widening gap between Tesla’s public safety narrative and the concerns of the people being asked to approve the technology.
The issue strikes at the heart of Tesla’s push into one of the world’s most tightly regulated auto markets. Europe is not simply waiting for a software update; its regulators are actively questioning whether the numbers Tesla has shown them accurately represent the system’s real-world performance.
The investigation, led by Reuters and corroborated by other outlets, found that Tesla’s widely publicized FSD safety statistics rely on what multiple researchers described as “deeply flawed methodology” . When Tesla CEO Elon Musk and other executives say FSD is up to ten times safer than a human driver, they are citing internal data that independent experts say compares two fundamentally different things.
The most damning detail came from Tesla’s own workforce. Reuters interviewed nine former Tesla data labelers—workers responsible for training the AI that powers FSD—who said they do not trust the technology to drive them personally . A former self-driving engineer and 11 traffic-safety researchers were also interviewed, painting what Electrek called a “damning picture of the gap between Tesla’s safety marketing and the reality of its autonomous driving program”
.
Regulatory bodies in Sweden and the Netherlands were the recipients of the data now in question. The figures were part of Tesla’s formal application to get FSD (Supervised) approved for public-road use across Europe, a process that began when Tesla approached the Dutch road authority RDW in late 2024 .
The response from European regulators has been unusually pointed. Internal emails obtained by Reuters from multiple national vehicle authorities reveal significant, ongoing skepticism toward FSD .
Three specific safety concerns appear repeatedly in the correspondence :
The Dutch regulator RDW—despite being the first to grant national approval—has been careful to stress that FSD remains a driver-assistance system. Its approval terms require the driver to stay fully alert and ready to take control at any moment, explicitly forbidding phone use or reading while the system is active .
Independent experts were more blunt. The data Tesla presented to Sweden and the Netherlands “could constitute misleading advertising,” according to traffic-safety researchers cited by Reuters .
As of mid-June 2026, four EU member states have granted national approval for Tesla’s Full Self-Driving (Supervised) on public roads :
A fifth country, Belgium, is reported to have approved FSD on June 10, 2026, bringing the total to five of the EU’s 27 member states—covering approximately 8.9% of the bloc’s population .
A much longer list of countries has FSD pending regulatory approval. Tesla’s Vice President of AI has confirmed that at least 12 countries are awaiting sign-off, including Germany, France, Italy, Spain, Portugal, Ireland, Luxembourg, Norway, Sweden, and the United Kingdom .
Even with multiple national approvals in hand, Tesla’s path to continent-wide authorization remains far from straightforward. Several structural and political obstacles stand in the way.
The UN R-171 mismatch. Tesla’s EU pathway depends on compliance with UN Regulation No. 171, a technical standard originally written for more predictable, constrained driver-assistance systems. The mix of behaviors Tesla wants FSD (Supervised) to perform—hands-off system-initiated lane changes, for instance—does not map cleanly onto the existing rulebook . The company needs exemptions through Article 39 clauses for behaviors that are still unregulated or tightly restricted in Europe, which gives national regulators a significant gatekeeping role
.
Country-by-country recognition is slow. Each EU nation must individually recognize the Dutch type approval or grant its own exemption. While the initial expectation was that major markets like Germany and France would act within 4 to 8 weeks of the Dutch green light, they have not done so as of mid-June .
The TCMV vote is the real prize—and it has no timeline. For truly binding EU-wide approval, the matter must go before the Technical Committee on Motor Vehicles (TCMV) for a formal vote. The Netherlands has already presented its testing data to the committee , but no date has been set for a vote, and political appetites among member states are visibly split
.
Sustained regulator skepticism. Those internal emails obtained by Reuters are not from a single skeptical authority but from several, including bodies in Sweden and Finland . The concerns about speeding, icy roads, and driver attention are not cosmetic; they go to whether the system can safely operate across Europe’s highly varied driving environments.
Timeline risk. Tesla has publicly targeted an EU-wide rollout in Q2-Q3 2026 , but that timeline now looks optimistic. The regulator friction, the lack of action from continental heavyweights Germany and France, and the absence of a scheduled TCMV vote all suggest the process will stretch beyond the company’s stated window.
Tesla’s European FSD story in mid-2026 is one of incremental, hard-won national victories against a backdrop of data-driven doubt and institutional caution. Four small-to-midsize EU countries have said yes; the continent’s largest car markets have not. The Reuters investigation has given regulators documented, specific reasons to slow-walk their decisions, and the fractured national-by-national approval process means Tesla cannot simply declare a European launch by fiat.
The next inflection point is likely the WP.29 session of the UNECE World Forum in late June 2026, where a draft global technical regulation on automated driving is up for adoption . If passed, it could eventually simplify Tesla’s approval path—but even optimistic observers acknowledge that near-term deployment still depends on winning over the national authorities that remain on the fence.
Studio Global AI
Use this topic as a starting point for a fresh source-backed answer, then compare citations before you share it.
Tesla submitted self generated safety data to regulators in Sweden and the Netherlands which independent traffic safety experts say is built on flawed methodology and could constitute misleading advertising.
Tesla submitted self generated safety data to regulators in Sweden and the Netherlands which independent traffic safety experts say is built on flawed methodology and could constitute misleading advertising. Tesla's own data labelers—the workers who train the AI system—do not trust the technology enough to use it themselves, according to interviews in the Reuters report.
Despite securing national approvals in the Netherlands, Lithuania, Estonia, and Denmark, Tesla's Q2 Q3 2026 target for EU wide clearance is now viewed as ambitious due to the piecemeal national approval process, an up...