Autonomy Without Accountability

The $32 Company Trading at $416

The number that frames everything about Tesla (TSLA) is the gap between two prices. The market values the stock at roughly $416 a share, a market capitalization near $1.56 trillion, and a trailing price-to-earnings ratio close to 400. Wealth Engine Pro's blended fair value model, which runs discounted cash flow, peer comparison, and earnings power against Tesla's actual reported financials, lands at $32.39. The market is paying about 13 times what the fundamentals support.

That gap is not an error in the model. It is the autonomy bet. Almost none of Tesla's current price reflects the cars it sells today. It reflects the business Tesla has promised to become: Full Self-Driving sold as recurring software, a robotaxi fleet earning revenue around the clock, and eventually a licensed driving stack other automakers pay to use. Strip that promise out and you are left with a car company whose 2025 revenue fell for the first time in its history, down 3% to $94.8 billion, with deliveries off 8.6%.

This article is not a forecast of where the stock goes next. It is about two specific developments from the last week of May 2026, both involving the same competitor, that cut directly at the autonomy promise carrying the other $384 of Tesla's share price. One attacks the price Tesla can charge for self-driving. The other attacks the cost structure underneath it. Neither shows up in a quarterly model yet. Both are the kind of thing that decides whether a 400 multiple ages well.

The Commoditization Leg

Start with price, because it is the simpler of the two moves. In China, BYD (OTC: BYDDY) sells its "God's Eye" driver-assistance system across its entire lineup, including models that start near $9,555. The company's founder, Wang Chuanfu, has framed the technology as no longer an unattainable luxury but standard equipment, comparing it to seat belts and airbags. The most capable LiDAR-equipped version, God's Eye B, is a one-time option priced around $1,770.

Tesla's comparable system in China, rebranded there as "Tesla Assisted Driving" and sold elsewhere as Full Self-Driving (Supervised), costs roughly $9,400 as a one-time purchase, with no subscription option in that market. In the United States, Tesla moved Full Self-Driving to subscription only in early 2026 at $99 a month. Either way, BYD is offering broadly comparable highway and urban driver assistance for about one-fifth of what Tesla charges. When BYD first put more than twenty God's Eye models on sale, analysts compared the pricing shock to what DeepSeek did to the AI sector, and shares of rival Chinese automakers fell on the news.

The scale behind that price is the part Tesla should find uncomfortable. BYD says it now has 3.15 million vehicles on the road with assisted driving active, generating up to 200 million kilometers of driving data per day. Fleet scale and data volume were, until recently, Tesla's signature advantage in the autonomy race. A rival is now matching the input that was supposed to be Tesla's moat and giving the output away close to cost.

This matters most in the one market Tesla has been counting on for its next leg of Full Self-Driving growth. Full Self-Driving is still not approved for sale in China, a market analysts have estimated could represent on the order of 2 million potential subscribers. Tesla is waiting to enter at a premium price into a market where the dominant local player has already made the same capability close to free. Software margins assume pricing power. Pricing power assumes the product is scarce. BYD is in the process of demonstrating that it is not.

The Liability Leg

The second development is the one the entire self-driving industry has spent a decade avoiding. On May 28, 2026, at its vehicle intelligence strategy event, BYD said it will assume full financial liability for at-fault accidents that occur while its God's Eye urban driving function is active in China, with no cap on the payout.

The terms are unusually direct. If a driver is using the urban navigate-on-autopilot function in compliance with regulations and is at fault in a crash, BYD will cover the direct economic losses the vehicle is liable for: repairs to the owner's car, third-party property damage, and personal injury. Owners do not have to buy a separate insurance product, and a claim does not raise their premium the following year. The coverage runs for one year from delivery, applies to the God's Eye A and B systems, extends to existing owners once they update their software, and is not restricted to the first or registered owner of the vehicle. Wang Chuanfu described it as taking on Level 3 and Level 4 liability while the system is still officially Level 2, a public statement of confidence in the technology.

BYD has done a version of this before. In July 2025 it attached a similar guarantee to its automated parking feature, and it says the pledge pushed actual usage of that feature from 21% to 93%. The new policy extends the same logic from parking to city driving, which BYD frames as a double guarantee. The guarantee is partly a marketing instrument: it sells hardware, and it drives the usage that feeds more data. But the mechanism does not change what it commits the company to. BYD is putting its own balance sheet behind the claim that its system works.

Tesla has never made that commitment. Its Full Self-Driving is a Level 2 system, and the responsibility sits entirely with the driver. Tesla's own owner's manual is explicit that the driver is responsible for the speed and control of the vehicle at all times, whether Full Self-Driving is engaged or not. When the system performs, it is marketed as nearly autonomous. When it fails, the driver was always in charge. BYD just took the opposite position in public.

Schrödinger's FSD

There is a name for the contradiction Tesla has been living inside. Critics have called it Schrödinger's Full Self-Driving: the car is driving itself right up until something goes wrong, at which point the human was always the one driving. The same software is autonomous enough to justify a premium and a trillion-dollar narrative, and supervised enough to keep every dollar of crash liability with the customer.

This is not a quirk of marketing. It is the load-bearing wall of the business model. The reason Tesla can sell Full Self-Driving today, expand a robotaxi service, and carry a 400 multiple is that it has captured the upside of autonomy while pushing the downside, legal and financial, onto drivers. Level 2 classification is what makes that possible.

The problem is that real autonomy cannot stay there. A system that makes the moment-to-moment driving decisions has to carry responsibility for them, which is precisely why Level 3 and above require the manufacturer to take on liability. The day Tesla delivers the unsupervised Full Self-Driving it has promised for years is the day it can no longer say the driver is responsible. At that point it inherits exactly the liability BYD just volunteered for. BYD is building that capability now, in public, with its own money on the line. Tesla is, for the moment, defending the opposite position in court.

The Safety Record

The liability question would be academic if Tesla's system rarely failed. The record says otherwise, and the record is documented rather than alleged.

In February 2026, a federal jury in Miami returned a $243 million judgment against Tesla over a fatal crash involving Autopilot, assigning the company roughly a third of the blame. Tesla fought the verdict, appealed, and lost. That single data point captures the difference between BYD's posture and Tesla's: one company is volunteering to pay, the other accepted responsibility only after a jury forced it.

Regulators are not finished either. In March 2026, the National Highway Traffic Safety Administration escalated its investigation into Full Self-Driving to an engineering analysis, the final stage before the agency can demand a recall, covering an estimated 3.2 million vehicles. The central finding is that the system's camera-based perception failed to detect common conditions like sun glare, fog, and airborne dust, and did not warn drivers that performance had degraded until immediately before impact. That probe began with four reduced-visibility crashes, one of which killed a pedestrian, and has expanded to nine incidents with additional crashes under review. It is one of three concurrent federal investigations into the system, the others covering traffic-law violations and Tesla's crash-reporting practices.

The pattern stretches back further. An earlier federal review logged hundreds of crashes in which Autopilot was alleged to be in use, a 2023 recall touched roughly 2 million vehicles over Autopilot safeguards, and testimony in a separate lawsuit indicated Tesla did not maintain crash records involving its driver-assistance technology until 2018. Tesla was also notably absent from a recent regulatory forum on autonomous driving guidelines that competing driverless developers attended.

There is context worth keeping in view, and the next section gives the other side its due. Reduced-visibility conditions impair human drivers too, and a single fatality triggering a federal probe is the kind of scrutiny any new safety technology attracts. But the pattern is hard to wave away: multiple open investigations, a recall-stage engineering analysis spanning millions of cars, and a nine-figure verdict. This is the record a company would carry into the moment it finally accepts liability. It is also the reason Tesla has every incentive not to.

What the Wealth Engine Scores Say

Before the valuation argument, here is what the Wealth Engine Pro platform's systematic scoring shows for this stock right now.

The platform rates Tesla as a Weak company trading at an Expensive price. The fair value estimate of $32.39 is essentially Tesla valued on what it reports today: the automotive and energy businesses, the margins, the balance sheet. At a recent price near $416, the market is paying roughly 13 times that estimate.

These scores are systematic. They evaluate companies based on reported financials, balance sheet quality, moat characteristics, and valuation models (discounted cash flow, peer comparison, earnings power). They measure what a company is today, not what it might become. That is by design: the scoring system is built to keep emotion and forward speculation out of the numbers.

That is also why the fair value sits at $32.39. The models can see the cars, the credits, and the energy storage. They cannot price the autonomy bet, because Full Self-Driving as high-margin recurring software, a scaled robotaxi network, and a licensed driving stack have not yet shown up as revenue and earnings in the filings. The market's $416 is almost entirely that bet.

In this case, the systematic scores and the editorial read point the same way. The platform says the current price is far above what today's fundamentals support. This analysis argues the forward story carrying that premium is being undercut from two directions at once. When the backward-looking numbers and the forward-looking argument agree, that convergence is worth paying attention to. Research Tesla yourself on the platform and decide which signal matters more for your situation.

What the Premium Is Pricing

Put the two developments back together and the question becomes specific. The roughly $384 of Tesla's share price that sits above its automotive fundamentals is paying for an autonomy business with two requirements. It needs pricing power: the ability to charge a premium for self-driving as recurring software and for robotaxi fares. And it needs to eventually deliver unsupervised autonomy at scale, the step that turns a car into a vehicle earning revenue around the clock.

The commoditization leg works against the first requirement. When the highest-volume electric-vehicle maker in the world offers comparable capability at a fifth of the price, in the market Tesla was counting on for growth, the premium narrows. The liability leg works against the second. The economics of a robotaxi fleet improve dramatically when a car earns fares for sixteen hours a day, but that math only arrives once the system is genuinely unsupervised, and that is the same moment Tesla stops being able to offload crash costs onto drivers. The cost it has avoided for a decade arrives with the revenue it has promised for a decade.

The current revenue is real but small relative to the promise. Tesla's Full Self-Driving subscriber base reached about 1.3 million in early 2026 after the shift to subscription pricing, and the deferred revenue tied to its software features is measured in the hundreds of millions, with about $941 million set to be recognized over the following year. The robotaxi service is live in a handful of cities with fleets still counted in single digits per metro, and meaningful robotaxi revenue remains, on the company's own pace, a 2027 story at the earliest. The promise is not fake. It is simply being priced today as if both requirements are already solved, at the precise moment a competitor is demonstrating that one is contestable and the other is expensive.

The Bull Case

The honest version of this thesis has to account for what could prove it wrong, and there is a real case.

Tesla's self-driving software is genuinely good and getting better. The most recent version has drawn praise even from outlets that scrutinize the company hardest, with one reviewer noting it is now smooth enough to make supervision feel complacent. Years of fleet data, more than ten billion driven miles, remain a real asset, and software improves with every update in a way hardware competitors cannot easily copy overnight.

BYD's liability guarantee, for all its symbolic weight, comes with limits. It is a China-only program, tied to compliant use of the urban function, capped at one year from delivery, and applied to a system that is still legally Level 2. It is as much a confidence play and a usage driver as a permanent change in how the system is classified. The price war BYD is running at home does not automatically cross into the United States or Europe, where regulation, tariffs, and brand dynamics differ.

The robotaxi optionality is also real. Sell-side estimates of the autonomous ride-hail market run into the trillions of dollars over the next decade. If Tesla scales unsupervised driving across a dozen metros, if Optimus reaches meaningful production, and if Full Self-Driving licensing closes with a major automaker, then a $32 fair value will look far too low in hindsight and the current price will look like an entry point. Tesla's first quarter of 2026 also showed the core business stabilizing, with revenue up roughly 16% year over year and an earnings beat. None of that is nothing.

The argument here is not that the autonomy business is worthless. It is that the autonomy business is being repriced in real time by events the market has not fully absorbed, and that a stock trading at 13 times its fundamental value has very little room for that repricing to go the wrong way.

The Bottom Line

For most of the last decade, the bull case for Tesla rested on two assumptions about self-driving that almost no one tested directly: that Tesla would own the technology, and that it could collect the revenue without ever quite owning the risk. In a single week, the largest electric-vehicle maker in the world challenged both. BYD made comparable self-driving cheap, and it put its own balance sheet behind the crashes, two things Tesla has spent years avoiding.

At Wealth Engine Pro, the question that matters is not what a company promises at a strategy event. It is what the numbers say and what the structure requires. The numbers say Tesla is a Weak company on its current fundamentals, priced at roughly thirteen times what those fundamentals support, with the difference resting almost entirely on an autonomy business it has not yet built. The structure says that business cannot generate the revenue it promises without one day accepting the liability it has spent a decade pushing onto its customers. A competitor just made both of those facts harder to ignore. That is not a verdict on where the stock trades tomorrow. It is a note on what the $384 premium is actually buying, and on how much of it now depends on assumptions a rival is busy disproving.