How did Theranos fool investors and regulators for over a decade?

For more than a decade, Theranos told the world it had built a machine that could run hundreds of blood tests from a single finger prick. Walgreens put “wellness centers” in stores. The board included Henry Kissinger, George Shultz, and James Mattis. At the peak, the company was valued at $9 billion and Elizabeth Holmes was the youngest self-made female billionaire. None of the technology actually worked. The answer to how is less about clever engineering and more about how a startup can hide behind board prestige, NDAs, and regulatory ambiguity for years if investors don’t push.

The first trick was access control on information. Theranos ran almost everything under strict non-disclosure agreements. The devices were kept in custom black boxes during partner visits. Investors were not allowed to see raw data, only summaries. The blood samples that partners like Walgreens actually sent to a lab were tested on conventional Siemens machines, not the Edison / miniLab devices — but the company kept the two pipelines separate so nobody could compare. When a scientist asked “show me the data,” the answer was almost always “you don’t have the clearance.”

The second trick was credentialing the board instead of the product. Holmes recruited a who’s-who of former secretaries of state, senators, and generals. The halo effect worked: a partner or investor who questioned the science would be reminded of who else had already signed off. By 2014 the board included Kissinger, Shultz, Mattis, Bill Frist, and George P. Shultz’s grandson Tyler Shultz — who later became the inside whistleblower who talked to journalist John Carreyrou of the Wall Street Journal. The retail partners and big-name investors weren’t doing diligence on the chemistry; they were trusting the résumés.

The third trick was hiding inside a regulated industry. Blood testing is overseen by the FDA and CMS, but Theranos positioned the miniLab as a “lab-developed test” that fell into a regulatory gray zone. For years, regulators deferred to the company, and the company told them the data was proprietary. That gray zone collapsed only after Carreyrou’s October 2015 WSJ exposé forced regulators to actually look. By 2018, Holmes and Sunny Balwani were indicted; by 2022, both were convicted of federal fraud, and Holmes received an 11-year sentence.

The lesson for an outside observer isn’t that the science was too clever to catch. It’s the opposite: the science was never shown. If you cannot get an investor-level look at the underlying data, the underlying chemistry, or independent validation from a domain expert, treat the company as unverified. For practical checks on smaller startups, see what red flags a small investor should look for in a health-tech startup, and for a related consumer-side fraud pattern, see how to protect your money from payment app scams.