Strategy raised $15 billion in preferred stock. AI designed the terms. The market doesn't care about your AI story. It only respects your exit strategy.
I’ve spent the last decade auditing smart contracts and trading through manias. The 2017 ICO boom taught me one thing: complex financial engineering often hides simple leverage. Strategy’s latest move—STRK and STRC preferred shares—is no exception. It’s a masterclass in turning Bitcoin volatility into a fixed-income product. But the underlying mechanics are fragile.
Let’s break down the structure before the narrative takes over.
Context: The Evolution of Strategy’s Bitcoin Acquisition Machine
Strategy (formerly MicroStrategy) holds over 840,000 BTC. That’s roughly $84 billion at current prices. To get there, they used every tool in the corporate finance toolkit: convertible bonds, at-the-market equity offerings, and now, preferred stock. The problem was scale. Convertible bonds had been used heavily. Equity dilution was becoming a concern. Saylor needed a new channel.
Enter AI. In an August 2025 podcast, Saylor described how he used AI to explore novel security designs. The result: STRK (fixed-rate convertible preferred) and STRC (floating-rate preferred). The AI didn’t write the legal docs—it generated the parameter space. The structure was executed by bankers and lawyers. But the narrative is powerful: “AI-designed securities” sounds like innovation. It’s actually regulatory arbitrage dressed in tech.
Core: The Financial Engineering of STRK and STRC
STRK carries a fixed 10% dividend and is convertible into common stock. STRC is the more interesting beast. It’s priced at $100 face value, with a floating dividend rate that adjusts based on market conditions. That adjustment mechanism is key. When Bitcoin rallies and demand for yield is high, Strategy can lower the dividend rate, reducing its cost of capital. When Bitcoin craters, they can raise the rate to keep investors from fleeing. In theory, it’s adaptive. In practice, it’s a credit instrument with a Bitcoin tail.
Audit the code, but trust the incentives. Here, the incentive is simple: Strategy needs continuous access to new capital to service existing dividends. The $15 billion raised (STRK: ~$10.5 billion, other preferred: ~$4 billion) is not a war chest. It’s ammunition for a rolling debt machine. The real source of repayment is Bitcoin price appreciation. If BTC appreciates faster than the blended cost of capital (~7-10% annualized), the structure works. If it doesn’t, the dividend obligations become a cash drain.
I’ve seen this pattern before. During DeFi Summer 2020, I built arbitrage bots for Uniswap vs. Sushiswap. The yield looked free until slippage and gas fees eroded it. Here, the yield looks safe until the underlying asset drops 30%.
Contrarian: The Retail Trap and the Hidden Leverage
The contrarian angle is that this is not a breakthrough. It’s a leveraged credit sale backed by Bitcoin. Saylor himself said, “We basically sold $15 billion of credit.” That’s honest. But the market treats it as a virtuous cycle: raise money, buy Bitcoin, raise more money. The feedback loop works in a bull market. In a bear market, it reverses.
Retail investors buying STRK/STRC at $100 face value think they’re getting a lower-volatility Bitcoin play. They’re not. They’re buying a corporate bond that depends on Strategy’s ability to keep borrowing. The dividend is paid from cash flow or new issuance. If the music stops, the dividend shrinks or the face value drops. The market doesn’t care about your thesis. It only respects your exit strategy.
Arbitrage isn’t just a trade; it’s a worldview. The true arbitrage here is between the narrative of “AI innovation” and the reality of financial engineering. Saylor is a brilliant storyteller. But the structure is fragile.
Takeaway: The Fragility of the Model
What happens if Bitcoin enters a three-year bear market? Strategy’s dividend obligations—$1.5 billion annually at a 10% blended rate—become a burden. New issuance dries up. The company may be forced to sell Bitcoin or dilute common shareholders. The stock price would collapse, and the preferred shares would trade at a discount to $100. The entire machine grinds to a halt.
This is not a prediction. It’s a risk analysis. The model is a “bull market accelerator, bear market amplifier.” The AI story is a distraction. The real question is: can Bitcoin sustain double-digit annual returns over the next decade? If yes, Strategy wins. If no, the preferred shareholders are left holding the bag.
Risk is invisible until it isn’t. I’ve seen this movie before. The 2022 Terra collapse taught me that algorithmic stability is a myth. This is not an algorithm; it’s a credit structure. But the same pattern applies: leverage works until it doesn’t.
Trust no one, verify everything. Verify the dividend coverage ratio. Verify the Bitcoin price floor needed to sustain the model. And remember: the market doesn’t care about your AI story. It only respects your exit strategy.