It is hard not to come across as a smug prick when you say, “I told you so.” Still, I am going to have to try. You see, just over a week ago, I made some bold, damning predictions about Tesla. I assumed that these would come true in a few months’ time, but it seems Musk has jumped the gun and proven me correct already. This effectively confirms that Musk is taking Tesla in a downward spiral of enshittification to line his own pocket. So, let me shed some light on the subject.
I made these predictions in the article “The Enshittification Of Tesla Has Been Turned Up To 11”, in which I explore why Musk is making FSD a subscription-only purchase from February. Ultimately, all of the conditions of Musk’s $1 trillion pay package have been phrased in such a manner that Musk can meet them without actually delivering any tangible growth, and I highly suspect that is what is going on here.
One of these conditions is that Tesla will “reach ten million active subscriptions to Tesla’s Full Self-Driving (FSD)”. This doesn’t measure how well Tesla’s self-driving business is doing at all because the definition is too loose, too unconstrained. Case in point: Musk removing the ability to buy FSD outright would nudge him closer to this goal without delivering any tangible real-world growth.
Especially when no one wants to buy the FSD system. In 2024, only 2% of drivers who tried FSD went on to purchase it. By the end of 2025, only 12% of the Tesla fleet had bought FSD, and FSD revenue in Q3 2025 was actually lower than in Q3 2024. We really shouldn’t be surprised by this when you look at Tesla’s FSD-powered Robotaxi service. Musk promised that it would cover half the US population by the end of 2025. Yet in 2026, it consists of just ten Robotaxis bumbling around Austin, Texas, at any one time. Not only that, but these Robotaxis crashed seven times over 250,000 miles, making their crash rate ten times higher than that of human drivers! And that is with a safety driver in the car, meaning that they weren’t truly autonomous. Despite this, Tesla’s Robotaxi recently began operating without a safety driver in the car, though it appears they have simply been transferred to a chaser car, making it appear more like smoke and mirrors than actual progress. No wonder no one wants to buy this system, as it is light-years away from being even remotely usable.
FSD is arguably the biggest automotive sales flop in history. Yes, even worse than the Cybertruck. Tesla has spent somewhere between $10 billion and $20 billion developing FSD and is expected to spend around $5 billion a year on development going forward. Yet FSD generates just a few hundred million dollars in revenue each year, and there are only marginal improvements in its performance and safety.
