
Analysis
3 ways to invest in Open AI
Disruption snapshot
Public investors can’t buy OpenAI directly. They must use Microsoft, private deals, or DXYZ. Each option has different liquidity, pricing risk, and structural complexity.
Winners: Microsoft and distributors with enterprise reach. Losers: investors who overpay for private shares or closed-end funds trading at steep premiums to asset value.
Watch DXYZ’s premium or discount to its net asset value. Also track Microsoft’s Azure AI revenue growth as proof that distribution is converting into cash.
The OpenAI trade is mostly a Microsoft (MSFT) trade in disguise.
That’s not boring, it’s the point.
Distribution beats novelty when the market gets crowded.
OpenAI is still private.
So the public market plays proxies, wrappers, and workarounds.
Some are sensible, some are pure adrenaline.
Your job is to know which one you’re holding.
Because “OpenAI exposure” can mean revenue share, private paper, or a fund markup.
Those behave differently when markets get ugly.
Via Microsoft
Microsoft is the cleanest public proxy for OpenAI.
It’s also the most diluted exposure.
That tradeoff is why it works.
Microsoft said in October 2025 that, after a recapitalization into OpenAI Group PBC, it held an investment valued around $135B, roughly 27% on an as-converted diluted basis.
That’s not the same thing as owning a simple block of common stock, and it won’t move tick for tick with OpenAI’s private valuation.
But it does give Microsoft economic participation plus something more important, distribution power.
OpenAI can invent, but Microsoft can ship.
Azure, Office, Windows, GitHub, and enterprise procurement are the rails that turn models into recurring revenue.
If OpenAI becomes as foundational as spreadsheets, Microsoft is positioned to collect rent.
Not just via its OpenAI stake, but via cloud consumption, copilots, and workload migration into Azure.
The upside is broad and boring, which is often the best kind.
The downside is you’re betting on execution across a huge company.
You’re also exposed to any renegotiation risk inside the partnership, plus regulation, plus the possibility that AI margins compress as models commoditize.
Still, if your goal is “own the distribution layer of frontier AI,” Microsoft is the most liquid way to do it.
Via private-market access
Private exposure is the closest you can get to “owning OpenAI.”
It’s also the easiest place to overpay.
And it’s where hype does the most damage.
OpenAI is not publicly listed, and it has used nonstandard structures over time, including a capped-profit setup and, as of October 2025, a PBC structure under OpenAI Group PBC with a nonprofit foundation alongside it.
That complexity matters because your rights in private deals can vary a lot.
Common, preferred, SAFEs, SPVs, and forward contracts can all claim “exposure” while behaving very differently.
What private access usually looks like in practice.
Secondary marketplaces, brokered blocks, tender offers, or SPVs offered to accredited investors.
Sometimes it’s employee liquidity, sometimes it’s late-stage funds recycling positions.
Two reality checks before you chase it.
First, pricing can run far ahead of fundamentals, especially in hot rounds.
Second, liquidity is discretionary, meaning you might not be able to sell when you want.
And yes, OpenAI fundraising is still evolving fast.
Nvidia was nearing a $30B investment as part of a massive OpenAI raise, with the round reportedly valuing OpenAI in the hundreds of billions.
That kind of headline can pull secondary prices into the stratosphere before terms are fully understood.
Private access is for investors who can read documents, size positions small, and live without liquidity.
If that isn’t you, treat this lane like venture capital, not like a “stock pick.”
Via DXYZ ETF
DXYZ is the loudest “public wrapper” story in this category.
But it isn’t an ETF.
It trades like a meme-able closed-end vehicle with real risks.
Destiny Tech100 trades on the NYSE under ticker DXYZ and is structured as a closed-end fund style vehicle, not an ETF.
The pitch is simple.
Give public investors a way to access a curated basket of late-stage private tech companies.
The disruption upside is obvious.
If DXYZ holds economic exposure to elite private names, it can become a bridge product for a world where everything valuable stays private longer.
That’s a legit trend, and it’s not going away.
The catch is mechanics.
Closed-end products can trade at big premiums or discounts to what the underlying portfolio is worth, because supply and demand set the price.
So you can be “right” on OpenAI and still lose money if you bought the wrapper too expensively.
Also, portfolio exposure can be indirect.
Some vehicles use SPVs or other structures to get economic exposure to private companies, which can introduce fees, complexity, and tracking error versus the headline name.
So if you buy DXYZ, you’re also buying governance, valuation methodology, and wrapper behavior.
Risks and the simple decision
This is a proxy game, not a purity contest.
Each route has a different failure mode.
Pick the one you can actually hold through volatility.
If you want liquidity and durability, Microsoft is the “rails” bet.
If you want closeness and can handle lockups, private-market access is the sharpest tool.
If you want a public wrapper on private tech, DXYZ is a high-voltage option, but you must respect premiums, discounts, and structure.
Recommended Articles



