Quick Thoughts: The AI Conglomerates
IPOs, Money Moves and Big changes!
Warning: Outdated content may here in this article! Reader’s discretion is advised.
This is a tour into the world that is on everyone’s tongue. If anything this goes deeper from the perspective of someone who hasn’t had the resources of an office and deeper into someone who uses AI for research and development. Needless to say, the only assistant I have is me, myself, and AI.
Throughout my writing you’ll notice I use Nano Banana for most of my thumbnails. Since I’m not exactly a graphic designer, and I’m working with a budget of exactly $0, I am outsourcing what I can. I build something quick that fits the content and move on. I’m doing the same here, so we’ll see what it comes up with. I also use Gemini and Claude to help with research. Kind of a win win for me until these companies say otherwise.
Now that you have a short primer into some of my personal use cases, let’s talk about the elephants in the room. Three major IPOs have dropped and are dropping and each one carries enough hype to move money in ways we haven’t seen in years. IPO season is here, and it is stronger than it’s been in a long time.
SpaceX (SPCX): The Elon Premium Is Here and Real
Let’s start with SpaceX because this one is going to live on the market’s tongue for a long time, possibly forever. Honestly, it already has.
SpaceX priced at $135 per share on June 11, began trading on Nasdaq under $SPCX on June 12, and raised $75 billion at a $1.75 trillion valuation, the largest IPO in stock market history. Larger than Saudi Aramco. Larger than anything that came before it. I know, not brand new news to us.
The numbers behind that headline are complicated though. SpaceX posted $18.7 billion in revenue for 2025, a 33% year-over-year jump, but also reported a $4.94 billion GAAP net loss for the year, with losses accelerating to $4.28 billion in Q1 2026 alone. The culprit is largely x.AI, which burned over $6 billion in 2025 after being absorbed into SpaceX in February 2026.
Don’t get me wrong, I am genuinely excited about SpaceX being public. That said, excitement and immediately buying are two different things for me. My instinct is to wait until the lockup periods for early shareholders expire and those first investors start taking chips off the table. That is when price discovery gets closer to real. Once the dust settles, I’ll start to run proper DCFs and decide if the Elon Premium is worth paying. The premium will be there regardless. The question is how much of it is justified by Starlink’s $6 billion in EBITDA and how much is pure narrative.
Morningstar has already called fair value at approximately $780 billion, roughly 55% below the IPO price. This gap is worth sitting with.
x.AI Is Now Built Into the Rocket
Since SpaceX absorbed x.AI in February 2026, the two are no longer separate businesses. The AI infrastructure they are building, anchored by the Colossus 1 supercomputer cluster, is now being commercialized directly. This means SpaceX is selling compute to companies like Anthropic and other compute intensive companies. The Terrestrial Compute Monetization move is a reality and creates a new revenue stream that did not exist roughly over a year ago.
Does this distract from the Mars mission? In Musk’s framework, no. The argument is that selling compute funds the rockets. But SpaceX investors are now implicitly betting on the AI infrastructure race too, whether they realize it or not.
The question I keep coming back to: how deeply does x.AI’s direction shape SpaceX’s operational priorities, and what happens to the stock when those two ambitions start pulling in different directions?
OpenAI: The ChatGPT Wants Its Turn at the Window
OpenAI filed confidentially for IPO on June 8, 2026, one week after Anthropic and right as SpaceX was running its roadshow. Needless to say the timing is a bit uncanny.
The current private valuation sits at $852 billion post-money from a $122 billion funding round in March 2026. The IPO target range is being reported at $730 billion to north of $1 trillion. Goldman Sachs and Morgan Stanley are leading the deal here as well, with an original September 2026 being their initial listing window.
Now as of recently, OpenAI is watching the pull back on SpaceX and is looking more into 2027 for their IPO. OpenAI itself was unusually candid in saying the timing “may be a while” because there are things they would rather do as a private company first.
This can be viewed as a tell for their financial reality or a true strategic move. Likely both.
The financials are eye-opening. $25 billion in annualized revenue run rate, growing at roughly 3x year-over-year, but burning approximately $27 billion per year in cash. One analysis found OpenAI is losing $1.22 for every dollar it earns. They target profitability for 2029 to 2030.
OpenAI is a generational brand with 900 million weekly users on ChatGPT. The product is sticky, the enterprise business is scaling, and Codex is competing directly in the developer workflow market. I would note that buying at 34 to 40x revenue before audited financials are publicly available is more of a prayer and not necessarily an investment thesis. The S-1, when it goes public roughly 15 days before the roadshow, will be the most important document in the AI market this year. Read it before you touch this one.
Anthropic: The Antithesis of B2C. . . B2B
Full disclosure: I am using Claude, built by Anthropic, to assist with research for this very article.
Anthropic filed confidentially with the SEC on June 1, 2026, targeting an October Nasdaq listing. The most recent private valuation is $965 billion after a $65 billion Series H round in May, which means Anthropic is currently valued higher than OpenAI on paper. The annualized revenue run rate crossed $47 billion in May 2026, up from roughly $1 billion in December 2024. That is not a typo. 47x revenue growth in 18 months.
The business is API-first, enterprise-focused, and anchored by Claude. Eight of the Fortune 10 are reportedly customers. Claude Code alone is generating $2.5 billion in annualized billings. Amazon and Google have both committed billions in compute and distribution deals, which is great for scale and also creates a dependency risk worth reading carefully in the eventual S-1.
The above strategy here is genuinely interesting. Anthropic sells AI safety as a feature. Though, they were placed on a Department of War supply chain risk list in February 2026 after refusing to allow Claude to be used for mass surveillance and autonomous weaponry. Then their latest models — Fable 5 and Mythos 5 — were restricted under the U.S. Commerce Department due to concerns about their ability to bypass ID software vulnerabilities and coordinate cyber attacks. Now that ban has recently been lifted but the decision to do so will show up in the IPO as both a risk factor and, depending on who is reading the prospectus, a brand differentiator.
Of the three, Anthropic is the one I would spend the most time studying before the IPO window opens. Revenue growth at that pace, with clear enterprise traction and a defined product moat in Claude Code, tells a fundamentally different story. But compute costs of $19 billion in 2026 spend and a path to cash-flow profitability not until 2028 mean the valuation is still faith-based until the S-1 drops. October is the window. Watch for the public filing.
Google Gemini: Not an IPO But Capitalizing on the Trend
The fourth elephant is one that I can’t and won’t stop thinking about. I think of this as completely rooted in the fact that they have total, unfettered access to the web’s core data layers. Which means Google is able to actively tap into their internal pipelines, combined with the intense engineering focus of Google DeepMind under Demis Hassabis, makes them an absolute powerhouse.
I know this sounds more bullish on Gemini than the others, but my reasoning is simple: Google will Google. They are uniquely capable of funding and “burning” cash without letting their core business fundamentals wither. While startup hyperscalers are relying heavily on speculative equity rounds, Google is flexing its massive corporate credit muscle.
To fund an unprecedented $180 billion to $190 billion capex budget for 2026 alone, Alphabet tapped global debt markets for a staggering $32 billion bond sale in early 2026. Over the past 12 months, they’ve quietly stacked over $85 billion in fresh debt across multiple currencies just to secure the concrete, steel, and custom TPUs (Tensor Processing Units) needed to run Gemini at scale. And just to keep the balance sheet pretty, they tacked on a massive $80 billion equity raise in June, anchored by a rare $10 billion vote of confidence from Berkshire Hathaway. Let me reiterate. BERKSHIRE HATHAWAY!
In a war of attrition where the cost of entry is a mountain of hardware, the legacy search giant can stay solvent and liquid longer than almost any other conglomerate on Earth.
The Wild Take: Three Trillion-Dollar Listings Walk Into a Bar
Here is what nobody is saying loudly enough. SpaceX, OpenAI, and Anthropic combined are seeking to raise more capital in the next two years than all U.S. public listings combined since 2022. This is huge when thinking about how it may affect liquidity and the broader financial economy.
Now, I expect retail money is going to pile into all three with varying degrees of mania. The hype is real, the stories are compelling, and the names are culturally impossible to ignore. But the lockup expirations, the cash burn reality checks, and the very first quarterly earnings calls as public entities are where actual price discovery happens.
Across all three, watch the doors when the lockups end. That is when the narrative finally meets the math.