Remount RoundupIssue 3
Wall Street just repriced the robot
A thought experiment - JPMorgan just tripled its Tesla target on the promise of machines that earn money on their own. The question nobody's asking: who actually gets to own that income?

The number that moved this week
JPMorgan has spent years as one of the loudest Tesla bears on Wall Street. This week they upgraded the stock and moved their price target from $145 to $475. That’s not a tweak. It’s a 227% revision - roughly a trillion dollars of “this is worth more than we thought,” in a single Friday note.
Here’s what’s strange about it: almost nothing about the company changed between the $145 call and the $475 call. Same cars, same factories, same roads. What changed is what the bank decided to count. They started pricing the robotaxis, the Optimus robots, and the self-driving software as real revenue instead of optional upside. By their own math, those three things - machines that earn money without a human doing the work - make up about half of where they think Tesla’s revenue goes over the next five years.
Sit with that. The single biggest re-rating of the year wasn’t about selling more cars. It was Wall Street finally admitting that a machine can go to work on its own and bring back income - and that the income is worth paying up for today, years before it fully arrives.
So here’s the thought experiment I keep running. When the market decides an autonomous machine is “real revenue,” who actually captures that?
Right now, one group: public shareholders. If you want a piece of the re-rating, you buy the stock. But think about what you’re actually buying. A share of Tesla is a sliver of a roughly $1.5 trillion company - cars, energy, insurance, robots, software, all of it - priced north of a hundred times forward earnings and diluted across billions of shares. You are not buying a robotaxi that earns $7,000 a year. You’re buying the average of everything Tesla does, and the income from any one car vanishes into that average.
There’s another way to own the same idea, and it’s almost the opposite. Instead of a thin slice of the entire company at a nosebleed multiple, you own the actual income-producing thing - the machine itself - at roughly the cost of the machine. When the world reprices “autonomous machines that earn money” the way JPMorgan just did, the cleanest way to catch that is to own one of the machines, not a fragment of the megacap built on the bet.
And notice the date on JPMorgan’s own target: December 2027. That lines up almost exactly with when Tesla has said regular people will be able to add their own cars to the robotaxi fleet. The bank is telling its clients the value is real and arriving by 2027. The same year, the door to owning the actual asset is supposed to open.
I’m not going to tell you which machine, or which company, or how to do it - that’s not what this note is for. The point is just this: the market spent this week deciding that machines doing autonomous work are worth a fortune. The only question left is whether that fortune belongs to the people who own the machines, or only to the people who already own the index.
— Jonas
Autonomy roundup - what you may have missed
A humanoid robot just worked a 200-hour shift - and never clocked out. This week Figure’s robots ran a continuous ~200-hour package-sorting line, tagging in to swap batteries so the work never stopped - fully autonomous on the company’s Helix neural net, no human at the controls, ~250,000 packages handled. Robotaxis aren’t the only machines starting to do the labor on their own; the warehouse version showed up this week too. The ownership question is the same for both.
🔗 Interesting EngineeringTesla stretched its Austin robotaxi map across the entire metro this week - Pflugerville, Manor, the I-35 corridor, the airport. The coverage map got big fast. The actual unsupervised fleet doing the driving is still small (independent trackers put it in the dozens). Coverage is running ahead of the cars - which is exactly what a land-grab looks like before the ramp.
🔗 ElectrekMeanwhile Waymo is doing ~500,000 paid driverless rides a week - and openly targeting a million by year-end. Driverless ride-hail isn’t a demo anymore; it’s a half-million-rides-a-week business that’s still doubling. The category JPMorgan just started pricing is already running a real meter.
🔗 TechCrunchThe two clocks are converging on 2027. JPMorgan dated its target December 2027. Tesla’s own guidance puts the owner “add your car to the fleet” program around the same time. The financial timeline and the ownership timeline are landing in the same year - which is the whole reason to get ready now, not then. 🔗 Tesla 8-K
Close
Reply and tell me where this breaks (or doesn’t) - I read every one.
And if you want to run the math yourself - what one of these machines actually earns, under your own assumptions - the modeler at remount.co/modeler lets you change every input.
You’re getting this because you signed up at remount.co. This is a thought experiment, not financial advice and not an offer of securities. Reply anytime or write hello@remount.co.