Platformonomics TGIF #49: April 26, 2024

European Union Designated an Advanced Persistent Threat

Q1 CAPEX Scorecard

The CAPEX numbers are rolling in:

• Google: $12.01B +91%, new all-time-high!
• Meta: $6.72B, -5%, but strong CAPEX guidance sinks stock!
• Microsoft: $14B +79%, new all-time-high!

Amazon reports next week.

Q1 Google CAPEX

Good trend — Google may finally have awoken:

Guidance (implies CAPEX could surpass $50 billion in 2024):

Looking ahead, we expect quarterly CapEx throughout the year to be roughly at or above the Q1 level

…most nearly all, I should say, of the CapEx was in our technical infrastructure. We expect that our investment in office facilities will be about less than 10% of the total CapEx in 2024, roughly flat with our CapEx in 2023, but is still there.

Q1 Microsoft CAPEX

Comparison with Google (an even better trend):

Notable:

Currently, near-term AI demand is a bit higher than our available capacity.

Guidance:

We expect capital expenditures to increase materially on a sequential basis driven by cloud and AI infrastructure investments. As a reminder, there can be normal quarterly spend variability in the timing of our cloud infrastructure buildout and the timing of finance leases. We continue to bring capacity online as we scale our AI investments with growing demand.

Therefore, we now expect full-year FY24 operating margins to be up over 2 points year-over-year even with our cloud and AI investments, the impact from the Activision acquisition, and the headwind from the change in useful lives last year.

To scale to meet the growing demand signal for our cloud and AI products, we expect FY25 capital expenditures to be higher than FY24. These expenditures over the course of the next year are dependent on demand signals and adoption of our services, so we will manage that signal thru the year. We will also continue to prioritize operating leverage and therefore, we expect FY25 operating margins to be down only about one point year-over-year, even with our significant cloud and AI investments as well as a full year of impact from the Activision acquisition.

Impressive:

Free cash flow grew faster (+21%) than earnings (+20%), despite the big bump in CAPEX.

Q1 Meta CAPEX

We haven’t historically tracked Meta because they’re not a cloud provider. But they are one of the biggest CAPEX spenders ($28 billion in 2023) and are accelerating that spend for generative AI. Huge props to Meta for tanking their stock 20% by upping their CAPEX guidance by $5 billion for 2024! We revere CAPEX-first companies!!!

Meta didn’t do a very good job explaining how they will make money from their increasing CAPEX investments. As described in a recent Zuckerberg interview, Meta is much more comfortable spending now and figuring it out later than Wall Street. But makes one wonder if Meta doesn’t get into model hosting for third parties in the future, making them more of a cloud competitor.

One explanation for the planned CAPEX bump:

Guidance:

We anticipate our full year 2024 capital expenditures will be in the range of $35 billion to $40 billion, increased from our prior range of $30 billion to $37 billion as we continue to accelerate our infrastructure investments to support our AI roadmap. While we are not providing guidance for years beyond 2024, we expect CapEx will continue to increase next year as we invest aggressively to support our ambitious AI research and product development efforts.

Notable — talking about CAPEX and energy as peers:

As we’re scaling CapEx and energy expenses for AI…

Meanwhile, at the Front of the GPU Queue

Hashicorp’s Acquisition by IBM

Some quick thoughts:

• Congratulations to all my friends at Hashi. $6.4 billion!
• The consolidation game is afoot and will continue — we still have too many sub-scale software companies.
• I feel slightly responsible as I suggested Hashi was a better acquisition than Red Hat five years ago. I also at that time expressed skepticism about IBM’s bombastic claim that acquiring Red Hat “changes everything about the cloud market”. Narrator: “it didn’t”.
• Hopefully Hashi can avoid the fate of other IBM acquisitions (company slogan: Where software goes to die). We will however reserve the headline “IBM Makes a Hash of Hashicorp” should it be needed in the future.
• The most interesting question is does IBM revert Hashi’s recent license change and try to undo the OpenTofu fork. IBM has been on its own quest to tighten up access to ostensibly open source RHEL so will be interesting to see how they play this.

And, just to point out the obvious, a Hashicorp acquisition doesn’t address IBM’s fundamental issues:

Platformonomics ClownWatch™: IBM

The OG CAPEX clown nailed the quarter with CAPEX declining 21% to a barely discernible $361 million (about 2.5 days of Microsoft CAPEX spend), while continuing to pretend they are an AI company (and they’re already pretending Hashi is an AI thing). IBM retains its position at the top of the ClownWatch™ list.

Platformonomics ClownWatch™: Tesla

Elon says Tesla should be valued as an “AI robotics company” and not a car company. People have thoughts on what that would mean financially:

But if we turn the all-seeing-eye of CAPEX to Tesla, we find:

• The claimed spend of $1 billion in “AI infrastructure CAPEX” is a suspiciously round number. Almost IBM-esque.
• But it is also Dr Evil-esque relative to what others are spending on AI infrastructure. Even noted CAPEX clown Oracle spends more than that on CAPEX.
• Telsa spent $2.8 billion in total CAPEX in Q1 (and we’re under the impression they have manufacturing plants and stuff). The previous four quarters were all over $2 billion so the $1 billion claim looks more like marketing than a big leap in investment.
• Tesla has never spent more than $9 billion in CAPEX annually.
• Telsa’s CAPEX is less than 3% of revenue, which is about what Microsoft and Oracle used to spend when they were pure software companies. Google and Microsoft now spend double digits of revenue. UPDATE: that 3% number was erroneously quarterly CAPEX over annual revenue. Tesla’s CAPEX as a percentage of revenue this quarter was 13%. But again, they have manufacturing plants and stuff.

So I am delighted to announce Tesla has been placed on Negative Watch. Further disparities between their AI rhetoric and their cold, hard CAPEX spend may result in the company being designated a full-fledged AI Clown.

And it looks like xAI may shortly be joining the list ($6 billion doesn’t take you very far in the frontier AI world):

Software Migration Alert: DarkTrace

When private equity comes amalgamating, it is time to start migrating. Especially when it comes to cybersecurity…

McKinsey: It Might Just Be Happening

Criminal convictions can lead to the death penalty.

Oracle Moves to Nashville

In the future, every city will be Oracle headquarters for 15 minutes.

And we have an exclusive photo of the entire Oracle Cloud headquarters region being migrating to Nashville:

Cloud Repatriation is Going Just Great: Continued Radio Silence

Interesting If True

So You Want to Build an AI Company: Episode 3

Their administrative bloat is your opportunity.