Our unboxed process for Cybercab assembles vehicle modules in parallel & frames the car in one step at the end.
This streamlines automation & reduces the line size by half, unlocking much greater production capacity.
The first real revolution in automotive manufacturing in over a century
Driving around Austin, Texas with my 5-year-old and 8-year-old when we spotted a driverless Tesla Cybercab in the wild.
So naturally… we started chasing it.
My kids grabbed the phone and filmed the whole thing.
This is literally their footage.
And then it hit me…
They may be two of the youngest kids in the world to ever see a driverless Cybercab operating on public roads.
Something they would not see today if they were at school today.
I don’t think they fully understand how big this moment is yet.
They just see a cool car driving itself.
But this is the world they’re going to grow up in.
A world where cars drive themselves, Robotaxis pick you up, and transportation looks completely different from what Daddy grew up with.
To me, this feels like the future.
To them, it might just feel normal.
One day they’ll look back at this video and realize they were there when that future was just beginning.
SpaceX and Tesla are each building 100GW/year of solar production capacity as fast as possible, but natural gas will still be needed to supplement and bootstrap solar for several years.
The limiting factor for nat gas turbine production is casting the blades & vanes. By doing in-house casting at SpaceX, we can accelerate nat gas turbines coming online by up to 18 months, which is a profound game-changer.
think about what you hear in early stage funding (for equity capital). Venture won't even listen to you unless you say AI.
That is now pushing downstream into the public equity markets (which people will claim is just equity investors losing their heads per usual) but also into the debt markets, as chips and datacenters use debt financing to build out the infrastructure.
Given the degree to which the market is compute starved the datacenter builders will still happily build at higher cost of financing (since their projects pencil so well), and the durability of the nvidia chip values means that so long as they're building nvidia they can collataralize debt against the chips, lowering cost of finance for even more balance-sheet tenuous operators.
I would expect that you will see this cut against high yield at first. After all the debt finance has to come from somewhere, but high yield financing generally sloshes around to where the marginal opportunity lies (and high yield investors are expecting some of their credits spin sideways on roll risk and general business idiosyncrasies.)
That compute will be presented and financed as a more stable more investment-grade-y asset class and so begin competing for funding against corporations and entities that aren't used to such a fierce level of competition is the surprising bit that I expect people aren't really dimensioning or anticipating.
As the annual capital requirements for datacenter build outs scales into the trillions, that becomes meaningful relative to the size of total global corporate issuance.
If AI infrastructure builders can tolerate a higher cost of funding because their projects underwrite better, where does that leave the incumbent stable businesses expecting to roll their credit at a more normal spread.
Of course the base rate matters as well, if real GDP growth inflects as we anticipate (and you can already see the leading indicators that this is going to occur) that should naturally drive long rates higher (even as technological deflation keeps inflation in check.)
Cross the two and very stable businesses that seemed to be chugging along fine could suddenly find themselves starved for inexpensive capital to a degree sufficient to destabilize their operating models.
Be very wary of businesses that claim they can live outside of disruption; we are at a pivot point, and the world is turning.