Tesla’s greatest threat is being boring: Liam Denning

Key topics:.Tesla Inc. EV sales slide as rivals erode its once-clear leadFSD shifts to subscription only, signaling commodification of autonomyNvidia Corp. threatens Tesla’s self-driving narrative with open AI systems.Sign up for your early morning brew of the BizNews Insider to keep you up to speed with the content that matters. The newsletter will land in your inbox at 5:30am weekdays. Register here.Support South Africa’s bastion of independent journalism, offering balanced insights on investments, business, and the political economy, by joining BizNews Premium. Register here.If you prefer WhatsApp for updates, sign up to the BizNews channel here..By Liam Denning.For a company that sells cars, Tesla Inc.’s $1.5 trillion valuation puts it in a league of its own. The stock is seemingly immune to bad news, even a growing cacophony of competitive threats. The more insidious challenge could be something more, for lack of a better word, boring.The most glaring example of Tesla losing its edge is the decline of its main business, electric vehicles. Sales of these have now fallen for two years in a row. China’s BYD Co. Ltd. overtook Tesla in sales of battery EVs in 2025. New figures from Cox Automotive show fourth-quarter sales in the US were down 15%, despite Tesla’s attempt at a post-EV tax credit reboot with stripped down trims of the Models 3 and Y.Now another, subtler signal has arrived in the form of a post on X by Chief Executive Officer Elon Musk:Tesla will stop selling FSD after Feb 14. FSD will only be available as a monthly subscription thereafter.FSD is Full Self Driving (Supervised), Tesla’s advanced driver assistance system that doesn’t fully self-drive the vehicle. This has long been available as an upgrade to buyers, currently costing $8,000. Since 2021, drivers have also been offered an alternative subscription model that costs $99 per month at the moment.As with so much else relating to Tesla, what makes this interesting are Musk’s own claims. In 2020, he posited that the FSD option (it lacked the “supervised” qualifier back then) would appreciate in value as the software improved, reaching $100,000 or more. As a rule, the value of technology, especially involving chips and attached to four wheels, tends to go the other way, and reasonably quickly. Better software could, in theory, change that. But Tesla’s own actions don’t indicate that happening.The price of FSD has fluctuated. It cost as little as $5,000 in 2019 before a series of price increases took it to a peak of $15,000. In the meantime, the subscription option was launched, and then the price began dropping to today’s level. Now, the option to buy FSD is going altogether, despite that being worth roughly 10 years of subscription revenue upfront to Tesla.So, not a vastly appreciating asset. More of a service where, as with pretty much any other, pricing gets adjusted according to the usual mix of relative capabilities, competition and market conditions.The sudden shift comes in the wake of Nvidia Corp. sending a few ripples through the Tesla-sphere at the recent Consumer Electronics Show in Las Vegas. The chip behemoth showcased Alpamayo, its open-source AI model for developing autonomous vehicles. Packaged with Nvidia’s existing hardware and software tools, it offers a potential off-the-shelf system for advanced driver assistance, and perhaps eventually fully autonomous driving, for automakers to install in their vehicles. Mercedes-Benz Group AG is set to deploy it in the US as soon as this quarter.Musk reacted by posting that Nvidia was essentially following in Tesla’s footsteps and, to a certain extent, it is. Nvidia’s machine-learning approach to autonomy is similar and it also faces a long road, literally, in proving its capabilities.The bigger point is that a tech giant with zero interest in making cars itself intends to offer something like the Android of autonomy to anyone willing to pay. This gives legacy automakers an opportunity to offer advanced driver assistance systems with less upfront capital expenditure, albeit at the cost of letting a growing source of vehicles’ value proposition leak away to a supplier that is valued at more than the entire global auto industry. This is a double-edged sword for Detroit, especially, with so much of its value concentrated in internal combustion engines.But it is potentially worse for Tesla by dint of the company’s reliance on expectations of dominance in self-driving cars to support so much of its valuation now that its EV business has hit the skids. The stock is, essentially, a story; a reflection of Tesla’s “narrative command” in the succinct phrasing of Alex Roy, an autonomy expert and co-founder of venture firm New Industry Management LLC. Such is Musk’s power to define the public consciousness around auto technology, including the very language used to describe it, that encroaching competition hasn’t dented Tesla’s 200-plus earnings multiple.Commodification, however, is competition’s creeping cousin. Almost a year on from BYD making advanced driver assistance standard across most models in China, Nvidia’s announcement adds another datapoint in what is actually a familiar pathway for vehicle features: The journey from pricey novelty to everyday tool. Remember that Autopilot, FSD’s less capable forebear, used to cost several thousand dollars before it became standard in Teslas. Switching solely to subscriptions with FSD — which can be canceled — after several price cuts, sends its own signal.Were Tesla’s offerings clearly unbeatable, this might not be such a threat. But, as I wrote here, there’s a curious lack of disclosure by Tesla when it comes to its apparently defining, and leading, technologies.Moreover, we are now only five months away from the first anniversary of Tesla launching its FSD-enabled robotaxis in Austin. To be clear, relative to Musk’s prior claims, that launch was farcical, and the company just missed yet-another deadline for removing the safety drivers in the vehicles’ front passenger seats.If the stately progress to date is maintained through June, will Tesla bulls grow at all uncomfortable with the notion that this company is truly leading the way? Roy points out that, barring something like a catastrophic crash, Tesla has other options for signaling progress on robotaxis and maintaining the narrative, such as expanding service to other cities with safety drivers still in place — and I have to agree that, based on Tesla’s cultish appeal, doing so could be enough to keep the stock aloft. Musk also has the option of leaning even harder into his robot narrative, of course, though competition is also evident there.Beyond the faithful, however, Tesla’s evident shortcomings versus its repeated pledges on driverless vehicles converge with increasing signs of not just competition but the beginnings of what might be termed normalization. It is curious to assume that a company which has clearly ceded unquestioned leadership and pricing power in its previous field, EVs, will somehow be immune from the same forces in its new ones.A company can succeed in some endeavor, any endeavor, and yet fail to meet bullish expectations as others muscle in and that endeavor loses its special aura. When valuing Tesla’s autonomy efforts, market share isn’t the only assumption that counts; price does, too. At such rarefied levels, the biggest threat may not be that of flaming out spectacularly, but rather the familiar drift toward banality..© 2026 Bloomberg L.P.

Tesla’s greatest threat is being boring: Liam Denning

Key topics:

Related Stories