With Tesla ($TSLA) well beyond the price target we set in March, we've been doing a ton of research behind the scenes trying to figure out the next move for the stock.
Automotive is becoming really competitive and Tesla is slowly shifting expectations regarding margins for their vehicles -- which is weighing on the stock. However, those headwinds appear to be lifting as Tesla finally gets its production cost-cutting underway.
But until we see significant cost-cutting in the actual earnings reports, it's tough to update our price target based on automotive alone given how cutthroat the competitive landscape is going to be.
But last month, Tesla passed a key milestone by turning on the first components of their massive DOJO supercomputer. Tesla has designed and built their own AI-specific chips to tackle their ambitions of developing a full self-driving software system. This is critical because now that this supercomputer exists, it looks like Tesla may actually pull off their absolutely ludicrous goal of 100 exaflops' worth of computing power by the end of 2024.
We'll explain how massive and insane that much computing power is later in the article, but the TLDR is that Tesla is on pace to becoming one of the top five organizations in the world in terms of computing power in the next 4 months.
So now analysts across Wall Street are raising their price targets to ludicrous levels because it looks like Tesla can really take a shot at solving one of the most difficult visual computing problems on the planet.
That's what Full Self-Driving (FSD) is. It's using software only to safely control cars on the road. It is such a difficult problem to pull off that there is still a decent enough chance that it's not possible. This is why, as we update our price target today, we do so with the underlying assumption that Tesla will fail in becoming a company that has software self-driving as a key feature.
This isn't a bearish report though. We simply are looking at the market opportunity that the Dojo supercomputer presents and are asserting that it doesn't matter if Tesla solves FSD any time soon. They are producing the most powerful visual computing system on the planet at a cost that 6 times cheaper and tons faster than any system Nvidia can produce.
Because of that, Tesla could service a huge slate of new industries and become a massively profitable B2B SaaS provider overnight. We haven't seen a company with an opportunity like this since Amazon launched AWS (and since AWS took over 70% of Amazon's profits).
Thanks to prudent design and being just a little ahead of the curve, Tesla is about to unlock years of profitable growth that will more than make-up for their vehicle margins staying compressed.
Tesla really is about to become a software business now that Dojo is real. But let's see just how powerful that can be and what it means for their stock price👇