The Exit Strategy Is Your Retirement Account

In late March 2026, the Fundrise Innovation Fund - a closed-end fund trading under the ticker VCX, holding small stakes in SpaceX, Anthropic, OpenAI, Anduril, and a handful of other pre-IPO companies - started trading on a stock exchange. Within four days, the price had been bid up more than 1,500% above the actual value of what the fund owns. Trading was halted twice because the price was moving too fast. Investors were paying sixteen dollars for every one dollar of real value.

Not because they didn't understand the mathematics.

Because they understood today's underlying economy principles (i.e. modern mathematics) - VCX was one of the only ways to own a piece of companies that have not gone public yet.

That is not investing. That is a line outside the club.

The club is about to open its doors. SpaceX, OpenAI, and Anthropic - with anticipated public valuations targeting $1.75 trillion, $1 trillion, and up to $900 billion respectively - are all planning to go public within the next twelve months. Together they expect to raise in the region of $200 billion from investors, against a market that handled $47 billion in IPOs all of last year.

The arithmetic does not balance.

Everyone involved knows this - just like all VCX shareholders knew.

Which is why the mechanics matter more than the headlines.

The Float

When a company goes public, it typically offers 15-25% of its shares on the open market. That portion - the float - is what investors can actually buy and sell. SpaceX is expected to offer approximately 3 - 5%. OpenAI and Anthropic are expected to use similar structures.

The consequence is well thought through - and previously tested. When you make a tiny amount of something available against enormous demand, the price reflects scarcity, not value. The investor is not paying for the company's underlying worth. The investor is paying for access to the company.

These are different things, and the financial outcome depends entirely on understanding which one is being purchased.

PitchBook analysts project that SpaceX stock will swing 20-30% on a single piece of news at this float level - roughly double Tesla's volatility (and Tesla is famously one of the most volatile large-cap stocks in existence). This is not an unintended consequence of the float structure.

It is the float structure operating as designed.

The Rules Changed in May

On May 1, 2026, Nasdaq's new "fast entry" index inclusion rules took effect. Under the previous framework, a company had to trade publicly for months - sometimes a full year - before it could be added to a major index. The logic was straightforward: give the market time to determine what a company is actually worth before requiring every index fund in the country to buy it.

Under the new rules, a company whose market capitalisation ranks within the top 40 of the Nasdaq-100 - roughly $100 billion - can be added after fifteen trading days. Three calendar weeks. S&P Global is currently evaluating equivalent fast-track rules for the S&P 500. FTSE Russell is considering similar changes to its indexes. Between the three, more than thirty trillion dollars in fund assets are connected to these benchmarks.

SpaceX reportedly made fast-track Nasdaq inclusion a condition of listing there rather than on the New York Stock Exchange. The company understood exactly what it was asking for: mandatory buying by every fund that tracks the Nasdaq-100, at whatever price the market is offering, during a period of artificially constrained supply. The rules that govern when your retirement fund must purchase a newly listed stock were rewritten, in the weeks before the listing, at the request of the company being listed.

If you hold a pension fund, a target-date retirement fund, or an index fund of any kind - and if it tracks any of the major benchmarks - you will own these stocks. Not because you chose to. Because the index rules require it.

The Lender of Last Resort

There is a financial reality beneath the narrative of AI dominance that received less coverage than it deserved.

OpenAI is expected to lose fourteen billion dollars in 2026. Its cash burn is projected to reach tens of billions per year by 2027. The company does not expect to achieve profitability until 2030. In January 2025, its CEO stood beside Donald Trump to announce Stargate - a $500 billion AI infrastructure project that would build data centres across the United States. Lenders were then asked to finance the construction. Lenders declined. The debt markets reviewed the financial statements and passed. The total projected compute spending was revised downward from $1.4 trillion to roughly $600 billion. Instead of owning infrastructure, OpenAI now rents compute capacity from Amazon and Google.

Read that sequence carefully.

The company announced a half-trillion-dollar infrastructure project at the White House. Traditional lenders, who are professionally required to read balance sheets, said no.

The public equity markets - which include peoples retirement savings - are the next door to knock on.

The IPO is not a celebration. It is a funding round. The public market is the lender of last resort, after every other source of capital has either been exhausted or has declined to participate.

Anthropic carries an additional structural risk. The company currently counts certain cloud computing credits from Amazon and Google as revenue - its cloud partners provide computing resources, and Anthropic books their value as income. Analysts estimate this accounting convention could represent billions in recognised revenue that would look materially different under alternative accounting treatment. If regulators require a change in this treatment before listing - as they are entitled to, and as an S-1 filing makes inevitable scrutiny - the revenue story shrinks. The S-1 is the first time the public sees the actual numbers. That moment will be instructive.

The Lock-Up Mathematics

When these companies list, the prices will almost certainly spike. Financial coverage will describe "stunning debuts" and "massive first-day gains". What that coverage will not foreground is that the spike is architectural: supply constrained to 3 - 5%, mandatory buying by index funds operating under rules that were changed weeks before the listing, institutional money managers required to own what the index owns or fall behind their benchmarks. The price reflects the structure. It does not reflect the company.

The question that matters is what happens at lock-up expiry.

Three to six months after the IPO, the 95-97% of the company that was not available begins flowing into the market. The sellers are the insiders who acquired their stakes at a fraction of the public price. A venture firm that invested in SpaceX at a $46 billion valuation in 2020 holds a 38x return at a $1.75 trillion public valuation. When the lock expires, that firm sells. That is the structure of venture capital: invest early, exit at the public market price, redeploy capital into the next cycle.

The buyers on the other side of that trade are the index funds. Which purchased at the supply-constrained IPO price. Which were required to purchase. Which will hold while the insiders who designed the structure execute their exit.

The gap between the price insiders paid and the price the index funds paid is a wealth transfer. There is no polite formulation that obscures the direction of that transfer, though "strong first-day performance" comes remarkably close.

The European Exposure

This is not only a problem for American 401(k) holders. EU pension funds track global indices. European retail investors in MSCI World ETFs will be exposed to these stocks within weeks of listing, through the same mandatory-buying mechanism, with the same absence of choice.

The EU's AI governance framework - the most structurally sophisticated in the world - has nothing to say about this. AI product regulation and financial market regulation are separate frameworks, maintained by separate institutions, operating at different velocities. ESMA has not issued guidance. EU member state financial supervisors have not published advisories. The European pension sector has no coordinated position.

The mechanism that transfers wealth from ordinary investors to company insiders operates at a pace that no regulatory apparatus, on either side of the Atlantic, was designed to match.

The encyclical asked the right question: in whose interest does this operate? The answer, at the moment of the IPO, is legible and precise. The answer at the moment of the lock-up expiry is equally precise, and considerably more uncomfortable for those whose savings are on the wrong side of it.

The Ideological Connection

The programme that animates this network is explicit in its rejection of democratic institutions. Its intellectual foundation - from Yarvin's Cathedral critique to Thiel's statement that freedom and democracy are incompatible - treats public oversight as an obstacle to be routed around, not a constraint to be engaged with. Regulatory frameworks are governance weapons deployed by the wrong side. Compliance is the cost of access to the market, not an obligation to the people the market affects.

The mechanism for realising the financial returns on this programme is the mandatory investment rules of democratic pension systems. The institutional architecture being philosophically opposed is the institutional architecture being financially relied upon. The anti-democratic intellectual programme is financed by the compulsory contributions of democratic citizens who did not choose to participate and, within the relevant timeline, have no practical mechanism to decline.

The Clause does not find this ironic. Irony implies an unintended consequence. This is the intended consequence. The rules that make your pension fund a mandatory buyer were changed at the listing company's request. The float structure that creates artificial scarcity was selected deliberately. The fast-track index inclusion timeline was negotiated as a condition of the listing venue choice. The sequence was designed by people who are very good at designing sequences.

What Happens Next

There are three plausible trajectories. The floats stay tiny and insiders hold for years, maintaining headline valuations that cannot be tested against real market depth. The public market reprices these companies once the S-1 filings reveal the actual financial statements to actual scrutiny. Or the forced-buying mechanism creates a self-reinforcing loop - mandatory buying drives prices up, higher prices validate the valuation, validated valuations attract more mandatory buying - until the loop breaks, as self-reinforcing pricing loops in financial markets historically do.

History gives self-reinforcing loops a name. It also documents their second act.

The line outside the club is your pension fund. The people inside the club are the insiders with the 38x returns and the expiring lock-ups. When the doors open, the arithmetic of who is selling to whom is not ambiguous.

Whether it should be legal is a governance question. Whether it was made possible by rule changes that were written to serve the companies doing the listing is a matter of public record. Whether the public record is receiving adequate attention before the IPO, rather than adequate coverage after, is the question that matters now.

The club opens in approximately twelve months. The queue is forming now, in May 2026.