Michael Saylor asked ChatGPT to help him solve a corporate finance problem. The chatbot’s answer ended up raising $15 billion.
Strategy Inc., the company formerly known as MicroStrategy, used AI-designed credit instruments to raise roughly $15 billion in capital, nearly all of it directed toward purchasing Bitcoin. Saylor disclosed the role of artificial intelligence in creating these hybrid securities during an appearance on The Diary of a CEO podcast, describing how the company turned to ChatGPT after hitting practical ceilings with traditional financing methods.
From convertible bonds to AI-engineered securities
Strategy had been funding its Bitcoin acquisitions primarily through convertible bonds for years. The company eventually ran into the natural limits of how much capital it could raise through conventional convertible debt, a wall that forced Saylor’s team to get creative.
Rather than hiring another investment bank to brainstorm, Saylor turned to ChatGPT. The AI helped design a new class of preferred stock instruments that blend characteristics of both debt and equity, a hybrid structure that hadn’t previously existed in quite that form.
The result was two new instruments: STRK, a convertible preferred stock, and STRC, which features a variable monthly dividend engineered to keep its market price stable near par value. One of these instruments alone raised approximately $2.5 billion, which Saylor described as the largest IPO of its kind at the time. Subsequent offerings stacked on top of that, eventually pushing the total haul to around $15 billion.
What $15 billion in Bitcoin buys
Strategy has deployed the proceeds into Bitcoin at an aggressive pace, bringing the company’s total holdings to 843,738 BTC as of late May 2026.
AI as corporate finance co-pilot
Traditional security design involves teams of lawyers, bankers, and structurers iterating over weeks or months. Saylor’s approach, at least as he described it, used ChatGPT as a starting point for ideation, generating structures that his team then refined and brought to market. The AI didn’t replace the lawyers and bankers. It gave them a blueprint they hadn’t considered.
The financial instruments still went through standard regulatory and legal review. But the conceptual architecture, the specific combination of convertibility features, dividend structures, and pricing mechanisms, originated from an AI model trained on the full corpus of financial literature and historical security designs.
What this means for institutional Bitcoin adoption
The variable dividend feature of STRC is particularly noteworthy. By adjusting payouts to stabilize the instrument’s price near par, Strategy created something that appeals to income-focused investors who might otherwise never touch a Bitcoin-adjacent security.
The risk, of course, is concentration. Strategy’s entire corporate identity now revolves around a single asset. If Bitcoin enters a prolonged downturn, the company’s ability to service its preferred stock obligations could come under pressure, regardless of how elegantly those instruments were designed.

