Anthropic is coming public, and almost none of us are getting an allocation.
Bloomberg reports the company could publicly file its IPO paperwork as soon as the end of this month, after confidentially filing back in June. Morgan Stanley, Goldman Sachs, and JPMorgan are running the deal. Anthropic is reportedly aiming to match or beat SpaceX’s record $86 billion raise from June, which would make this the largest IPO in American history.
I spent 13 years in equities at Goldman Sachs and Credit Suisse, so I know exactly how allocation works on a deal like this. Institutions paying the most in commissions get the shares. Private wealth clients with eight figures at the firm get some. Then a few retail investors get 50 shares so the platform can issue a press release about democratizing finance.
You are not getting a meaningful allocation. Neither am I.
So the actual question is not “how do I buy the Anthropic IPO.” The question is “what do I buy today that already owns Anthropic, before the IPO reprices everything.”
There are four realistic options. They are not close to equal.
Why Anthropic Is The Most Important IPO Since Google
Some numbers to frame why this matters.
Anthropic reported preliminary second quarter revenue of more than $11.5 billion, up from $787 million in the same quarter of 2025. That is roughly 14X growth in a single year. First quarter revenue was $4.73 billion, so it more than doubled sequentially. Annualized run rate hit about $65 billion by the end of July, ahead of OpenAI’s roughly $40 billion.
The company also reported positive adjusted operating income in Q2. An AI lab that is growing 14X and no longer setting money on fire is a genuinely rare animal.
Anthropic raised $65 billion at a $965 billion post-money valuation in May 2026. Secondary markets have since traded it at $1.2 trillion, and more recently north of $2 trillion. Reuters reports the company is projecting 2028 revenue of roughly $190 to $200 billion.
My own call: Anthropic IPOs at $2 trillion before year-end 2026, reaches $3 trillion by the end of 2027, and $5 trillion by the end of 2028. I could be wrong. But at $2 trillion against a $65 billion run rate, you are paying about 31X revenue for a business compounding at triple digits. Nvidia has traded there. So has Palantir, on far worse growth.
Option 1: Amazon (AMZN)
Amazon is the largest outside shareholder in Anthropic. It has put in about $13 billion and reportedly holds somewhere between 15% and 21% of the company.
The gains are already showing up in the financials. Amazon’s Q2 2026 net income jumped 245% year over year to $62.6 billion, and most of that came from marking up Anthropic. Its “other” net income line went from $1.1 billion to $53.4 billion. Two-thirds of the quarter’s pre-tax income came from a private company it does not operate.
At the end of June, Amazon carried the stake at $190.4 billion. At a $2 trillion Anthropic, it is worth north of $400 billion.
Here is the problem. Anthropic went from 3.4% of Amazon’s value at the end of March to 7.4% at the end of June. Even at a $2 trillion Anthropic IPO, the stake is only about 15% of Amazon.
So you are buying an e-commerce and cloud conglomerate and getting a side of Anthropic. If Anthropic triples, Amazon does not triple. Safest way to play this, and by far the least potent.
Option 2: SK Telecom (SKM)
Now this is a fun one. In August 2023, SK Telecom wrote a $100 million check into Anthropic at a $5 billion valuation. A Korean phone company made one of the best venture investments of the decade.
That stake started around 0.7% to 2% and has been diluted through subsequent rounds to an estimated 0.4% to 0.7%. At $965 billion, that is roughly $4 billion to $6.8 billion. Against a market cap of about $14 billion, the Anthropic position could be worth 30% to 45% of the entire company.
Morningstar reclassified SK Telecom as an AI stock. Its P/E went from 11X to 60X on the strength of one check.
The catch is that SK Telecom will not disclose the exact ownership percentage, citing confidentiality clauses. You are underwriting an estimate wrapped in a Korean holding company discount, with potential Korean tax and governance friction on any monetization. Fun trade. Not a clean one.
Option 3: Destiny Tech100 (DXYZ)
DXYZ is a closed-end fund holding roughly 30 private companies. Anthropic is its largest position at about 18% of the portfolio, alongside SpaceX at roughly 15%.
But look at the entry point. DXYZ bought $100 million of Anthropic exposure in February 2026 as part of a $127 million deployment, through a special purpose vehicle. In some cases those SPVs sit on top of additional layers of SPVs.
So DXYZ showed up around a $350 billion mark, through a wrapper, on top of a wrapper, with a 2.5% management fee. It is real exposure. It is just late, indirect, and expensive.
Option 4: VCX, The Fundrise Innovation Fund
This is where I have put my own money, and where I keep adding.
Fundrise bought its first Anthropic position in December 2023 for $8.5 million, when Anthropic was worth roughly $15 billion. Not through a triple-layered SPV in 2026 at $350 billion. Directly, in 2023, at 1.5% of today’s private mark.
Fundrise last officially disclosed Anthropic at 21% of the fund as of February 15, 2026, when Anthropic was marked around $350 billion. Anthropic has since raised at $965 billion. Even after accounting for dilution and for the rest of the portfolio appreciating, my estimate is that Anthropic is now roughly 30% of VCX’s net asset value. I walked through this math in detail in my VCX NAV estimate post, where I landed on $31 a share using a $1 trillion Anthropic. At $2 trillion, that number goes materially higher.
But the Anthropic concentration is only half the argument. Here is what else you get:
| Holding | Approx. % of VCX | Status |
|---|---|---|
| Anthropic | ~30% (my estimate) | IPO filing expected imminently |
| Databricks | ~18% | IPO candidate 2026-2027 |
| OpenAI | ~10% | IPO expected late 2026 or 2027 |
| Anduril | ~7% | Trading at ~2X last round in secondaries |
| Ramp | ~5% | Fintech, growing fast |
| SpaceX | ~5% | Already public as SPCX |
That is not an Anthropic proxy. That is a portfolio of the four or five most important private companies of this cycle, with the single best one as the anchor.
VCX trades around $40 with a market cap of about $1.16 billion. It fell from $575 at the March listing peak to an all-time low of $28.71 during the August 14 lockup unlock, the heaviest volume session in the fund’s history. I bought that dip. I wrote about what holding through that lockup actually felt like in my six-month lockup post, and about the SpaceX unlock read-through in this post.
I will not sell a single share below $60.
The Honest Risks
I am not going to pretend this is free money.
VCX is a closed-end fund with no creation and redemption mechanism, which means it can trade at a wild premium or a punishing discount to NAV for a long time. It did both this year. Citron ran a short attack on it. Short interest is up over 400% since March.
Anthropic’s marks are private estimates, not prices. The May round is 15 months stale by the time this thing lists. Anthropic has also been designated a supply chain threat by the Trump administration, which is a real regulatory overhang. Chinese open-source models are getting good and cheap.
And if the AI trade breaks, a fund with 30% in one private AI company does not go down 20%. It goes down 60%.
I am comfortable with that because I have a 10-year time horizon and because I would rather own the picks and shovels of the thing that might replace my children’s future jobs. Owning AI is my hedge against AI. I explained that logic in my private AI investing post.
My Take
If you want the safest exposure, buy Amazon. If you want a quirky deep-value trade, look at SK Telecom. If you want a diversified private basket where Anthropic happens to be the biggest slice, DXYZ works.
But if you want the most Anthropic per dollar, held the longest, at the lowest cost basis, alongside OpenAI, Databricks, and Anduril, VCX is the cleanest vehicle available to a normal person right now.
The window closes when the S-1 goes public and everyone can finally see the cap table. That could be in the next two weeks.
I would rather be early and slightly wrong than perfectly informed and completely out of position.
If you want a framework for how to build serious wealth through concentrated bets like this without blowing yourself up, pick up a copy of Millionaire Milestones: Simple Steps To Seven Figures, my USA Today bestseller with Portfolio Penguin. It walks through how to size positions, when to concentrate, and when to diversify.
Disclosure: Fundrise is a long-time sponsor of Financial Samurai, and I am a multiple six-figure investor in Fundrise products, including VCX. All valuation figures in this post are my own estimates based on publicly reported funding rounds and news reports, not official marks. Risk assets can and do lose money. This is not investment advice. Please invest at your own discretion.
Leave a Reply