SpaceX因收购Cursor及Grok增长而成为可信的AI竞争对手；Oppenheimer将目标价上调至280美元。

SpaceX was nowhere in AI just six months ago. Now it's an anthropic rival after the $60 billion cursor deal. And that is according to Oppenheimer. By the way, I talked to one of my long-standing developer friends who's developed everything under the sun, it seems. And he said that this cursor acquisition is not a joke. In long term, this is going to be extremely impressive. went on to praise Grock, uses Claude, um but also said, you know, Grock is is really incredible and has made some serious strides. Was very impressed with what I was able to do and was doubly impressed that someone who's not a traditional developer. Actually chose a command line interface that gives you diffs and developer terms. Didn't understand why that was the case. Turns out I'm learning a shitload about development and that is helping me do what I need to do. Sorry for swearing.

Oppenheimer analyst Timothy Heran says Space Exploration Technology Corporation or SpaceX, ticker SPCX, has gone from an artificial intelligence also uh ran to an incredible to a credible anthropic rival within months, crediting its $60 billion cursor acquisition and expanding compute infrastructure. Quote, "The company's doing an incredible job with AI." End quote. Haran told CNBC's Power Lunch host Kelly Evans and Brian Sullivan. Quote, he goes on to say, six months ago, their AI business was almost, people thought, dying. The large language models of Grock were nowhere. End quote. He called Curser an absolute gamecher. Curser, of course, officially joined SpaceX on August 14th. Iran says Curser's agentic coding activity gives SpaceX data on quote what works, what doesn't work, and all of the logic behind that end quote. And that helps improve Grock and other applications.

He said SpaceX is targeting a 100red billion revenue run rate by year end with about 70% tied to AI and could reach 120 billion to 130 billion next year. He also says, quote, I think Grockbot is about to go viral. End quote.

And by the way, um I some of the stuff that I watch at night has to do with AI, what's on the horizon, uh what's new, how I can adopt that to my own use cases and all of that. And I'll tell you something. The amount of videos that I see about Grockbot literally running entire businesses pretty much on their own makes me think that I'm actually behind and I should have my own business run by a whole swarm of AI agents. To me, it actually seems like this Grockbot uh virality, if that's the right word, it actually seems like that's starting already.

Ran had said that and well and also said people working on Grock 5 believe it will be transformational. Now I have Grock 4.6 right now and I can tell you that the jump from 4.5 to 4.6 was immediately noticeable. Like immediately noticeable. I can't even imagine what Grock 5 has in store for us.

Oppenheimer reinforced that thesis on Wednesday, maintaining its outperform rating and raising SpaceX's price target to 280 from 250, implying nearly a 99% upside from around 141. The firm said SpaceX's vertically integrated AI platform combines proprietary data capital and NVIDIA GPUs and rapid infrastructure deployment.

Oppenheimer raised long-term revenue estimates roughly 10% and said Nvidia's next generation Reuben chips could pay for themselves within a year and estimated SpaceX could capture 100% of revenue from software such as Cursor and Grock versus roughly 50% when it acts as a wholesale AI infrastructure provider.

SpaceX already leases substantial compute to Anthropic and Alphabet. That's ticker G O G and G O G L. An anthropics deal includes 1.25 billion in monthly payments through May of 2029, subject to a 90-day termination provision.

Quote, at a $2 trillion valuation, they really have to grow into that. End quote. This is again from Heran who argued that 300 billion to 400 billion of revenue within two to three years could support a 3 to4 trillion market cap. Quote, SpaceX really is now a competitor to Anthropic. End quote. And he went on to say, ' Horan, that is. It really wasn't one three, four, five months ago, and I think the world is waking up to that. End quote.

I'm old enough to remember when I was criticized saying that creating my bot with Grock meant that I was going to be a failure and that I could do so much more with Claude. Seems that I was a tad early on that. I'll wait and do my victory lap when it's all when it's all said and done.