Martin Shkreli: Big Pharma's Blind Spot, Betting Against Quantum & What Prison Taught Him

Martin Shkreli says he saw an inefficient market in drug rights: large pharmaceutical companies sometimes sold assets cheaply even when they had potential value. He says he bought a drug from Bristol Myers for $2 million and that the company later reached a value of $6 billion. He also says he repositioned sparsentan, originally developed for hypertension, for FSGS. In his account, sparsentan is FDA-approved and the first approved drug for a disease affecting roughly 30,000 people. The episode introduces Shkreli as a symbol of “Capitalistic Greed,” citing the 5,000% increase in the price of a life-saving drug and his seven years in prison.

A shortage of people able to vet drug assets led Shkreli to start a small natural-language-processing company in 2015, he says. Its aim was to help analyze 40 million published medical papers, addressing the challenge of finding promising opportunities across a large body of research. He distinguishes this work from Royalty Pharma’s model: rather than developing drugs, he describes Royalty Pharma as buying drug cash flows at a discount.

Shkreli rejects the idea that quantum computing is simply faster general-purpose computing. He says its distinctive advantage is polynomial-time algorithms for integer factorization and discrete logarithms, while most workloads remain better suited to conventional machines. He claims quantum computers operate at hundreds or thousands of hertz, lack quantum RAM, and cannot outperform NVIDIA machines on most tasks. He and his wife, he says, could not use any quantum computer they tried to multiply 300 by 200. That skepticism has not protected them financially: they lost millions of dollars shorting quantum-computing companies.

He questions the business case for Google and IBM spending hundreds of millions of dollars a year on quantum technology when its payoff may be far off, while acknowledging that the field could still be at an early stage. Shkreli and his wife are investigating analog optical computing and might start a company. He cites a Microsoft paper arguing that light can perform matrix multiplication effectively for free and instantaneously, but says nonlinearity and storage remain challenges. The possibility interests him without resolving whether optical systems can become practical computing products.

Shkreli says photonic computing could make H100, B100, and B200 processors obsolete within 20 years, but he is not bearish on semiconductors and says he is currently long semiconductor stocks. More efficient computing could hurt NVIDIA, while Jevons’ paradox could instead increase demand as computing gets cheaper. He says it is uncertain whether demand for compute can grow a millionfold, leaving the effect of greater efficiency on chip demand unresolved.

Shkreli says he began working at Jim Cramer’s hedge fund at 16, where the approach involved buying and selling throughout the day and returned 23% a year net of fees. Reflecting on prison, he says refusing to accept that he was not the world’s best hedge-fund manager—and to recognize his strengths as an entrepreneur—put him there. He says prison taught him that people share basic needs and feelings, and that becoming fixated on an identity such as “great trader” can be detrimental.