The $9 Billion Blood Test Lie: How Theranos Fooled Silicon Valley

Transcript

Thanksgiving Day, 2013. One California lab. One twenty-two-year-old, watching the same test fail again. A patient’s blood sits in front of her, and she says out loud that she will not send this result. Then a colleague walks in, deletes the ugly numbers, and calls them outliers. The company she works for is worth nine billion dollars, built on more than four hundred million in venture money. Its founder, Elizabeth Holmes, is about to be crowned the world’s youngest self-made female billionaire. Her board is stacked with Henry Kissinger, George Shultz, James Mattis, and not one medical expert. And the miracle machine? It was handing out inaccurate results while most tests ran on ordinary equipment anyone could buy. Every expert who understood blood had told her it was impossible, from the very beginning. So why did the money keep coming, and who was ever going to be brave enough to say the machines did not work?

Palo Alto, California, 2003. A Stanford University undergraduate sits across from a medicine professor named Phyllis Gardner and describes a machine that does not exist yet. The pitch is simple and enormous: take a few droplets of blood from the tip of a finger, and pull vast amounts of medical data out of them. Gardner listens, then tells her flatly that she does not think the idea is going to work. She explains that what the student is describing is impossible. Several other expert medical professors tell her the same thing. The student’s name is Elizabeth Anne Holmes, and she is nineteen years old. She does not relent.

To understand why she did not, you have to look at where she came from. Holmes was born in Washington, D.C., in February 1984. Her father was a vice president at a subsidiary of Enron; her mother had worked as a Congressional committee staffer. One of her paternal great-great-great-grandfathers was Charles Louis Fleischmann, a Jewish-Hungarian immigrant who founded Fleischmann’s Yeast in the United States. A family friend, Joseph Fuisz, later put it bluntly: the Holmes family was very proud of that yeast empire, and the parents yearned for the days when the family was one of the richest in America. He said Elizabeth channeled that, from a young age. She grew up in a house that remembered being rich.

She graduated from St. John’s School in Houston, in the United States, where she got interested in computer programming and says she started her first business selling C++ compilers to Chinese universities. Her parents arranged Mandarin tutoring at home, and partway through high school she started attending Stanford’s summer Mandarin program. In 2002 she enrolled at Stanford to study chemical engineering and worked as a student researcher in the School of Engineering. After her freshman year she worked in a laboratory at the Genome Institute of Singapore, testing for the SARS coronavirus by collecting blood samples with syringes. That is the detail worth holding on to. She has described her fear of needles as a motivation for wanting to run blood tests on tiny amounts of blood. She filed her first patent application, on a wearable drug-delivery patch, in 2003.

So she went looking for someone who would say yes. She found him inside her own engineering school. Channing Robertson was a Stanford chemical engineering professor who had joined the faculty in 1970 and served as senior associate dean in the School of Engineering. He had let Holmes work in a research laboratory alongside PhD graduate students, and he was swayed by what she believed her fledgling technology could do. When the medical experts said no, her advisor and dean said yes. Robertson helped her start the company, became its first board member, and gave up his academic tenure to work there. He also brought venture capitalists to meet her. A quote of his ended up on Holmes’s desk: that you start to realize you are looking in the eyes of another Bill Gates, or Steve Jobs.

In March 2004 she dropped out of Stanford’s School of Engineering and used her tuition money as seed funding. The company she founded in Palo Alto was first called Real-Time Cures. Then she changed the name, because she had decided that too many people were skeptical of the word cure. The new name was Theranos, a mash-up of therapy and diagnosis. Think about that for a second. Before there was a working device, there was a marketing decision about which word made people nervous.

And the money came anyway. By December 2004, Holmes had raised 6 million US dollars for a company barely a year old, run by a twenty-year-old with no degree. Her stated mission was to democratize healthcare, to make blood tests cheaper, more convenient and more accessible to ordinary consumers. It is a mission almost nobody would argue with. That may be exactly why so few people argued. Every expert who understood blood had told her it was impossible, so why did the money keep coming?

Picture a company with no website. No press releases, no launch, no product page, nothing for a journalist to call about. That was Theranos for its first decade: Elizabeth Holmes ran it in stealth mode, out of Palo Alto, in the United States. And the money loved it. By the end of 2010, the company had pulled in more than 92 million US dollars in venture capital. Start from the 6 million dollars she had raised by December 2004, and that is roughly 14 million dollars a year flowing into a firm the public could not even look up.

Around 2009, the board made a move that should have changed everything. Directors threatened to replace Holmes with someone more experienced. She had an answer ready. She promised to bring in a seasoned executive of her own choosing, and the man she named was Ramesh Balwani, who went by Sunny. He joined Theranos that year and took over the company’s day-to-day operations as president. What the board did not know was that Holmes and Balwani had been in a romantic relationship, and that they had met years earlier, when she was an 18-year-old high school senior on a Stanford Mandarin trip to Beijing, in China. He was 37 at the time, married, and 19 years older than her.

So who was this man being handed a medical laboratory? Balwani had worked in the 1990s at Lotus Development and at Microsoft, where his job was sales. He claimed to have written thousands of lines of code, but investigators could not verify it, and Microsoft managers asked about him could not remember him at all. His big score came in July 2000, when he sold his shares in a software company called Commerce One and walked away with nearly 40 million US dollars, just before the firm collapsed and the dot-com bubble burst. He later went to business school at Berkeley, then spent four years in a Stanford computer science program and dropped out in 2008. He had no training in biological sciences. None in medical devices either.

And there were no medical experts on the Theranos board to notice the gap. Former employees described Balwani as overbearing and uncompromising, so worried about industrial espionage that it verged on paranoia. Inside the company he reached for technical words he did not appear to understand. He once misheard the robotics term end effector as endofactor, a word that does not exist, and repeated it through an entire meeting. Colleagues later slipped Endofactor into a PowerPoint slide as a prank, and he did not notice. This was the same man who said the Edison blood testing device would rank up there with the discovery of antibiotics.

Together, Holmes and Balwani built what employees called a dysfunctional culture of secrecy and fear. Staff said anyone who raised concerns or objections was usually marginalised or fired. Erika Cheung, who joined from the University of California, Berkeley, said you had to sign a non-disclosure agreement before you even sat down for the interview. Employees were told not to put the company’s name on their LinkedIn profiles, and to write private biotech company instead. Teams were deliberately siloed so they could not talk to each other. Cheung described Balwani as a dictator and a fearmonger who was always yelling, and said Holmes had her own version of it, telling anyone who pushed back that they were not a team player.

There is one more thing the investors were never told. The chief executive and the president of Theranos were a couple, and that relationship was not disclosed to the people writing the cheques. It lasted more than a decade, through nearly all of the company’s history. So the lab was run by a salesman with no medical training, reporting to a founder he was secretly living with, inside a company nobody outside could inspect. All they needed now was a set of faces the world would trust on sight. So who exactly was watching the lab that was supposed to save lives, and why would the most famous statesmen in America agree to lend it their names?

July 2011. A meeting is set between Elizabeth Holmes and George Shultz, the former United States Secretary of State. It lasts two hours. By the end of it, Shultz has agreed to join the board of directors of Theranos. Two hours, one signature, and the whole character of the company changes. Over the next three years, Holmes was credited with assembling what was called the most illustrious board in United States corporate history. The names arrived one after another: Henry Kissinger, William Perry, James Mattis, all of them former or future cabinet secretaries. Former Senator Sam Nunn sat there too.

Look closely at that list and something is missing. Not one medical expert. That absence mattered, because the man running the labs day to day, Sunny Balwani, had no training in biological sciences or medical devices at all. Inside the company he used technical terms that colleagues believed he did not understand, trying to sound more knowledgeable than he was. He once called the Edison blood testing device an invention that would sit way up there with the discovery of antibiotics. Nobody on that board was qualified to tell him he was wrong.

But the statesmen did something money alone could not buy. They made the company look safe. Erika Cheung, who joined in 2013 straight out of the University of California, Berkeley, later said the board was part of why the job looked like a dream: Kissinger, Nunn, Shultz. By the time she arrived, Theranos had raised more than 92 million dollars. And the retail world was already opening its doors.

In 2012, the supermarket chain Safeway committed 350 million United States dollars to retrofitting 800 of its locations with clinics offering in-store blood tests. Do the arithmetic and that is roughly 437,000 dollars per store, spent on rooms for a machine no outsider had ever verified. Then came the missed deadlines, and questionable results from a trial clinic at Safeway’s own corporate offices. That deal would eventually be terminated in 2015. But by then the bigger partner was already in.

In September 2013, Theranos finally came out of stealth mode. Until that point there had been no press releases and no company website at all. The announcement was a partnership with the pharmacy chain Walgreens, putting in-store blood sample collection centres into more than 40 locations. Real patients. Real needles. Real results going to real doctors. Cheung, who started around that time, watched patient samples begin rolling in from the Walgreens centre in Palo Alto, California. What happened to those samples is the hinge on which this entire story turns.

The press fell in love. Eric Topol, editor-in-chief of Medscape, praised Holmes for what he called a phenomenal rebooting of laboratory medicine. In 2014 she appeared on the covers of Fortune, Forbes, The New York Times Style Magazine and Inc. Forbes named her the world’s youngest self-made female billionaire and ranked her number 110 on the Forbes 400. Theranos was valued at 9 billion dollars, on more than 400 million dollars of venture capital raised. By the end of that year, her name was on 18 United States patents and 66 foreign ones.

And the man who had started it all, her old Stanford dean Channing Robertson, was still there, vouching for her. He compared her to Leonardo da Vinci, to Mozart, to Einstein and Newton. He also kept quiet about what was happening inside the company, telling not even his wife. He was paid about 500,000 dollars a year, according to a lawyer who sued Theranos on behalf of former investors. That same lawyer described Robertson as the person in the world who would have known exactly the right questions to ask. With generals, diplomats and a Stanford professor all vouching for her, who inside that building would dare stand up and say the machines simply did not work?

Thanksgiving Day, 2013, in a lab in California, in the United States. The building is empty. One person is at the bench: Erika Cheung, twenty-two years old, a few months out of the University of California, Berkeley, with a degree in cell biology and linguistics. She has a patient’s blood sample in front of her, and before she can run it, she has to pass quality control. It fails. She resets the whole system and runs it again, and it fails again, and again. She calls the company help line and says plainly that she does not feel comfortable sending out this patient’s result.

Here is what happened next, and it is the moment the whole story turns on. A research associate came in, looked at the failing numbers, and deleted data points, telling her those were outliers she could get rid of. Cheung had found her dream job at a Berkeley careers fair earlier that year. By then Theranos had a board stocked with Henry Kissinger, former Senator Sam Nunn, and George P. Shultz. She was, in her own words, a starry-eyed twenty-two-year-old, and she genuinely admired Elizabeth Holmes as a role model. So she told herself the same thing a lot of young employees tell themselves. Maybe this was normal.

It was not normal. Patient samples were arriving from the Walgreens centre in Palo Alto, California, and the controls kept failing, night after night, once keeping Cheung at work until three in the morning. Untrained staff were making calls on results, and upper management had one instruction: get the results out, fast, at any cost. The machines used fragile glass tips that jammed and broke constantly, so Cheung put up error sheets to track every failure, and her colleagues filled them in. Then the sheets vanished. She says Ramesh Balwani was a dictator and a fearmonger who was always yelling, and that Holmes dismissed anyone who pushed back as not a team player. She cried at the office more times than she can count, and she says she has never cried at work since.

She was not alone at those lunch tables. Her closest friend in the building was Tyler Shultz, who had graduated from Stanford University in 2013 and started at Theranos just before she did. He had met Holmes in his grandfather’s living room, through George Shultz, the former Secretary of State who sat on the Theranos board. His grandfather encouraged him to ask for an internship, and it turned into a full-time job as a research engineer on the Edison device. Over roughly eight months, Shultz and Cheung and others found internal malpractice, manipulated data, false positives in syphilis tests, and falsified blood samples. Inside the company, they found, it was an open secret that the technology did not actually work. The two of them ate lunch together every day and traded what they were seeing, including the fact that patients were being handed results from tests that had failed.

On the eleventh of April, 2014, Tyler Shultz took his concerns straight to Elizabeth Holmes. Nothing changed. What he got instead was an email from Sunny Balwani, belittling him. He resigned. Cheung quit the same month, seven months after she started, after crying down the phone to her father, who told her she was a smart girl and he trusted her to figure it out. Shultz did not let it go, either: he reported Theranos to the New York State Department of Health for laboratory violations, using an alias to hide who he was.

Remember the cost of that phone call. Shultz would later face a lawsuit from Theranos and run up half a million dollars in legal fees alone. For years he believed private investigators were following him. Cheung says the company tried to intimidate her into silence; one day colleagues told her someone had been sitting in the parking lot all day, and she was handed a letter at an address in Palo Alto where she was only staying temporarily, an address even her own mother did not have. The letter accused her of disclosing trade secrets and threatened to sue. Two twenty-somethings against a nine-billion-dollar company and a former Secretary of State: who on earth was ever going to believe them?

The answer came from a newsroom in New York, in the United States. John Carreyrou was an investigative reporter at The Wall Street Journal, a man who had already shared two Pulitzer Prizes. In late 2015 he began a series of investigative articles on Theranos, the blood-testing startup founded by Elizabeth Holmes, questioning the company’s claim that it could run a wide range of lab tests from a tiny sample of blood taken from a finger prick. What set him off was not a leak from inside the company. It was a tip from a pathologist named Adam Clapper, and deep doubts published by a researcher, Eleftherios Diamandis. Carreyrou’s investigation was secret, and it ran for months. He spoke to ex-employee whistleblowers and he obtained company documents.

The academics had been circling for most of that year. That February, Stanford professor John Ioannidis wrote in the Journal of the American Medical Association that Theranos had published no peer-reviewed research in the medical literature at all. Think about what that means for a company then valued at nine billion dollars. Not one paper. Three months later, University of Toronto professor Eleftherios Diamandis, in Canada, analysed the technology and concluded that most of the company’s claims were exaggerated. Theranos had a nine-billion-dollar valuation and a scientific record of zero published studies. And someone inside the building was about to confirm exactly why.

Then Holmes’s lawyer, David Boies, learned the investigation existed. What followed was not a rebuttal. It was a campaign to stop Carreyrou publishing, built on legal and financial threats aimed at both the Journal and the whistleblowers, tied to the supposed release of trade secrets. Theranos sent lawyers after the sources in the story, including Tyler Shultz, to stop them talking to the press. Shultz would eventually run up half a million US dollars in legal fees alone. For years he believed he was being tailed by Theranos private investigators, and Carreyrou suspected the company watched them both for a year.

Then Holmes went over the reporter’s head. She turned to Rupert Murdoch, whose media empire owned The Wall Street Journal. Murdoch was not a neutral party. He had become the biggest investor in Theranos that year, with an injection of 125 million US dollars. Holmes asked him to kill the story. Murdoch refused, saying he trusted the paper’s editors to handle the matter fairly. The single largest shareholder in the company declined to protect his own investment.

That October, the bombshell landed. The Journal reported that the Edison device gave inaccurate results, and that Theranos had been running most of its testing on commercially available machines made by other companies. Machines anyone could buy. The whole promise, a drop of blood on a proprietary box, and the box was mostly theatre. Theranos called the allegations factually and scientifically erroneous. Walgreens, the retail partner, immediately suspended plans to expand its blood-testing centres.

That same evening, Holmes appeared on CNBC’s Mad Money. Jim Cramer told her the article was pretty brutal. Holmes answered that this is what happens when you work to change things: first they think you’re crazy, then they fight you, and then all of a sudden you change the world. She called the Journal a tabloid and promised the company would publish data on the accuracy of its tests. Meanwhile, former employees of a reputation firm said Theranos had hired them to quietly scrub mentions of the reporting from its Wikipedia page. And there was one performance that had already gone further than any television interview. When Holmes invited the sitting US Vice President, Joe Biden, to tour the facility, he praised what he saw, because she and Sunny Balwani had built him a fake lab to look at. When federal inspectors finally walked into the real laboratory, what would they find waiting for them?

The letter landed that January, and it used three words no laboratory ever wants to read. The Centers for Medicare and Medicaid Services had inspected the Theranos lab in Newark, California, and told the company that the facility posed immediate jeopardy to patient health and safety. The trigger was a test that helped doctors set the dose of warfarin, a blood thinner. Get that number wrong and a patient bleeds, or clots. This was no longer a fight about a magazine cover or a newspaper story. This was a federal regulator saying, in writing, that people could be harmed.

Walgreens moved fast. It suspended all Theranos tests coming out of the Newark lab and immediately paused its wellness services in Palo Alto, and Theranos agreed to stop testing at the Capital Blue Cross store in Enola, Pennsylvania. Elizabeth Holmes went on The Today Show and said she was devastated that the company had not caught and fixed the issues faster. She promised the lab would be rebuilt, with a new scientific and medical advisory board to help. The regulators were not waiting for the rebuild. That March they proposed banning Holmes from owning or operating a certified clinical laboratory for two years, because the problems in California still had not been fixed.

Then came the part that made the whole story physical. Theranos told regulators it was voiding every test result its Edison machines had produced in 2014 and 2015, along with some tests it had run on conventional machines. Two years of answers, wiped. Every one of those results had gone to a real person, and to the doctor who acted on it. The company’s own accounting, that June, was that about 1 percent of results from its proprietary machines had been voided or corrected. Within two weeks, Walgreens formally ended the partnership and began closing every remaining Theranos wellness centre.

In July the hammer fell properly. CMS revoked the company’s CLIA certificate, suspended its approval to receive Medicare and Medicaid payments, issued a civil monetary penalty, and barred its owners and operators from running a clinical laboratory for two years. Theranos said it would appeal. The Food and Drug Administration also ordered the company to stop using the Capillary Tube Nanotainer, one of its core inventions. Sunny Balwani had already left his post that May. The company that promised a lab in every pharmacy in America had lost the right to run one.

And then the number that everyone remembered simply evaporated. That June, Forbes revised Holmes’s personal net worth from 4.5 billion US dollars to nothing. Not a haircut. Zero. Paper had made her the wealthiest self-made woman in America, and paper took it back in a single revision. By October, Theranos announced it would shut its laboratory operations and wellness centres and cut about 40 percent of its workforce.

The last bill came from Arizona, where the company had grown fastest. The state alleged Theranos had sold 1.5 million blood tests to Arizonans while concealing or misrepresenting important facts about them. Theranos settled, agreeing to refund every Arizona resident who had taken a test, whether or not that person’s result had been voided or corrected, for a total of 4.65 million US dollars. Do the arithmetic and the refund works out at roughly three dollars per test sold in that state. Three dollars, for a result a doctor may have trusted with a life. Walgreens, meanwhile, sued in federal court in Delaware and originally sought 140 million US dollars in damages.

By the start of the following year, Theranos had laid off another 155 people and closed its last remaining blood-testing facility after failing a second major regulatory inspection. The labs were dark. The partners were gone. But refunds and revoked licences are civil matters, and nobody had yet been accused of a crime. The money was gone and the machines were switched off, but would anyone actually be charged with a crime?

That December, with the labs closed and the lawsuits stacking up, Theranos was reportedly on the verge of bankruptcy. Then money arrived from an unlikely place. Fortress Investment Group, a company wholly owned by SoftBank Group, loaned Theranos 100 million US dollars for 4 percent of the company. The loan was meant to keep the firm solvent into the following year. And it was secured by something specific. Not the buildings, not the machines that never worked, but the patents. That detail would matter more than anyone realised at the time.

Elizabeth Holmes had once been the youngest self-made female billionaire in the United States, on the strength of a 9 billion dollar valuation. Now her company survived on a loan against paperwork. At its height, Theranos had more than 800 employees. That April, the company filed a notice with the State of California saying it would permanently lay off 105 people, leaving fewer than two dozen behind. From 800 down to under 25. That is roughly 97 out of every 100 jobs gone. The building emptied while the legal machine finally got moving.

In March 2018, the U.S. Securities and Exchange Commission filed civil fraud charges against Holmes, Theranos and Ramesh “Sunny” Balwani. The regulator said they had raised more than 700 million US dollars from investors through an elaborate, years-long fraud with false or exaggerated statements about the technology, the business and the finances. One of the claims was that the technology had been used by the U.S. Department of Defense in combat situations. It had not. Another was a revenue stream of 100 million US dollars in 2014. The real figure for that year was 100,000 US dollars. So for every thousand dollars the company claimed it earned, it actually earned one.

Holmes settled the SEC case without admitting or denying wrongdoing. The terms tell you how far she had fallen. She paid a fine of 500,000 US dollars, returned 18.9 million shares to the company, gave up her voting control, and accepted a ten-year ban on serving as an officer or director of any public company. Balwani did not settle. He said he was innocent. He would get his day in court, and he would not enjoy it.

Then came the part that civil charges cannot do. In June 2018, after an investigation by the U.S. Attorney’s Office in San Francisco that had run for more than two years, a federal grand jury indicted Holmes and Balwani on nine counts of wire fraud and two counts of conspiracy to commit wire fraud. Prosecutors alleged two separate criminal schemes: one to defraud investors, another to defraud doctors and patients. Both pleaded not guilty. Holmes resigned as chief executive of Theranos after the indictment, but she stayed on as chairwoman of the board. Chairwoman of what, exactly, was becoming a fair question.

That summer the lights went out for good. Most of the remaining employees were let go on the last Friday of August. Days later, Theranos emailed its investors to say it would cease operations and hand its remaining cash and assets to creditors, after every effort to find a buyer had failed. The general counsel, David Taylor, was by then the chief executive, and he and a few support staff stayed on the payroll a little longer to close it down. Any equity investment in the company was made worthless by the shutdown. About a week later, Theranos transferred its patents to Fortress in exchange for its cash going to creditors. Fortress later used Theranos patents to set up a shell company called Labrador Diagnostics, which sued a maker of COVID-19 tests during the pandemic. The company was dead, and its inventions were suing people. The money was gone, the doors were locked, and two people faced prison, but would a jury call it fraud, or just a startup that failed?

San Jose, California. A federal courtroom. Elizabeth Holmes takes the witness stand in her own defense. She stayed there for seven days. That is almost unheard of in a white collar fraud trial, because every hour on that stand is an hour the government gets to cross-examine you. Her lawyers came from Williams and Connolly, an American firm known for defending white collar cases. The judge was Edward Davila, in the U.S. District Court for the Northern District of California. And the trial had already been pushed back more than a year, delayed by the COVID-19 pandemic and by Holmes’s pregnancy. So what does a founder say when the evidence is faked demonstrations and falsified validation reports?

She said she had been misled. Holmes testified that her own staff had told her things about the technology that were not true. And she pointed at the man who had run the company beside her. She told the court that Ramesh “Sunny” Balwani had held sway over her during their long romantic relationship, a relationship that was still going on while the alleged crimes happened. She said he was controlling, that he berated her, that he sexually abused her. She said Balwani wanted to kill the person she was and build a new Elizabeth. In court filings, Balwani and his ex-wife categorically denied the abuse allegations and called them false and inflammatory. And then came the sentence that undercut the whole defense. Holmes testified that Balwani had not forced her to make the false statements to investors, partners, journalists and directors.

January 3, 2022. The jury came back. Guilty on four counts tied to defrauding investors: three counts of wire fraud and one of conspiracy. Not guilty on four counts tied to defrauding patients. On three other investor counts the jury could not agree, the judge declared a mistrial, and prosecutors later dropped them. Read the split and you can see what twelve ordinary people decided. The evidence in the case laid out Holmes’s role in faked demonstrations, falsified validation reports, misleading claims about contracts and overstated financials. The product that actually sold was the story.

One person celebrated with champagne. Tyler Shultz, the young researcher who had reported the company under an alias, had been considered as a witness but was never called. He was in the overflow room during closing arguments, spotted by reporters, saying nothing until the verdict landed. His father, Alex Shultz, did take the stand, describing Holmes’s vengeance and manipulation against their family. Remember, Tyler had spent half a million US dollars in legal fees alone after Theranos sued him. And for years he believed private investigators were following him.

Balwani got his own trial. A judge had ordered the two co-defendants separated, so his case ran after hers. On July 7, 2022, the jury convicted him on all twelve counts. Not four. Twelve. That November, Holmes was sentenced to 11 years and 3 months in prison. In December, Balwani was sentenced to 12 years and 11 months, plus three years of probation. He surrendered on April 20, 2023, and went to a federal facility at Terminal Island in San Pedro, California. Holmes reported to the Federal Prison Camp in Bryan, Texas, on May 30, 2023.

Then came the number that was supposed to make the victims whole. In May 2023, Holmes and Balwani were ordered to pay 452 million US dollars in restitution, shared between them. That money was owed to twelve victims of the fraud. Investors had put more than 700 million US dollars into the company. So the restitution order covers roughly sixty cents of every dollar they raised. A court order, though, is just a piece of paper. Somebody still has to write the check. So how much of that 452 million US dollars have the victims actually seen?

Start with a retired financial planner in Texas, in the United States, checking his email. His name is Alan Eisenman. He put about 1.2 million US dollars of his family’s money into Theranos starting in 2006, which made him the smallest fish on a list of investors that included the Australian media mogul Rupert Murdoch. He testified at Elizabeth Holmes’s trial. Because he is a victim of a federal crime, the Justice Department has to keep him informed, and that Wednesday the notification system had news. Holmes would be moving to a halfway house in August 2027. Eisenman had one word for it. “Seems like a gift,” he told NBC News.

Here is how the arithmetic got there. Holmes was sentenced in 2022 to 11 years and 3 months, and she has been serving it at a minimum-security prison in Bryan, Texas. Then, that March, a judge cut the sentence from 135 months to 123 months. The reasoning was that Holmes had not personally caused substantial financial hardship to her victims. Twelve months, gone. A consultant named Walter Pavlo, who runs the firm Prisonology, said a mix of things could shorten her time behind bars and in the halfway house: changed sentencing guidelines, good conduct credit, and programs that shave off days when you finish them. Her scheduled release date is February 22, 2030.

The government fought the reduction. Prosecutors told the court that Holmes had not accepted responsibility for her crimes, that she had paid minimal restitution, and that she still had the opportunity to reoffend. And they pointed at something specific. Holmes, they said, had kept telling national media she intends to go back to a career in health care technology once she is out. She had also been advising her romantic partner on raising money for his startup, a company the filing described as having similarities to Theranos. That partner is Billy Evans, the hotel heir she married in 2019 and with whom she has two children.

Now put the numbers side by side. Holmes and Ramesh “Sunny” Balwani were ordered to pay twelve victims a total of 452 million US dollars. Eisenman and his wife, Sherrie, are owed almost 150,000 US dollars of that. That is roughly one three-thousandth of the total bill. He says he has not received a penny. “She was a liar back in 2006,” he told NBC News, “and she’s a liar today.” Eisenman wanted her to serve every day of the sentence, and he is against her getting any form of presidential clemency. She filed for a commutation last year, and as of that reporting it was still pending.

Balwani’s path runs on a different clock. He was convicted on all twelve counts in July 2022 and sentenced that December to 12 years and 11 months, plus three years of probation. He fought to stay out on appellate bail, lost, and surrendered on April 20, 2023. He is held at Federal Correctional Institution, Terminal Island, in San Pedro, California, in the United States. The Bureau of Prisons lists his expected release date as April 11, 2034.

And the two twenty-somethings who blew the whole thing open? Tyler Shultz was named to the Forbes 30 Under 30 for Healthcare in 2017, the year Forbes said he had shaped the biggest healthcare story of the year. That same year he founded Flux Biosciences, a company trying to put health diagnostics inside ordinary households, and in 2022 he founded another, The Healthyr Company, built around analysing blood samples. His grandfather, George Shultz, sided with Holmes first and stopped speaking to him for months. They reconciled before the old man died in 2021. Erika Cheung, who cried at her desk over quality controls that would not pass, went on to found a nonprofit called Ethics in Entrepreneurship, aimed at stopping the next Theranos before it starts. Shultz advises it. His story is now taught as a Harvard Business School case. If the sentence keeps shrinking and the ambition never does, what exactly is stopping the next Theranos?

So let’s answer the question we started with. Did the technology ever work? No. In 2016, Theranos voided two years of results from its Edison device. The Wall Street Journal had already reported that the company was running most of its tests on ordinary commercial blood machines built by other companies. Regulators found the Newark lab posed immediate jeopardy to patient health and safety. The lab certificate was revoked. The wellness centers closed. And in September 2018, Theranos told investors it would shut down and hand its remaining cash to creditors. Every equity investment in the company became worthless.

Why did the money keep coming for so long? Because credibility was borrowed, not earned. A board stacked with Henry Kissinger, George Shultz and James Mattis. Magazine covers from Fortune, Forbes and Inc. A Stanford professor who told a fund manager the technology was years ahead of rivals, shortly before that fund put in $96 million. Meanwhile, no peer-reviewed research had ever been published, as John Ioannidis pointed out in 2015. Nobody in the room demanded the data. It took a 22-year-old lab associate, a 23-year-old research engineer and one reporter with a tip to ask for proof. And the bill was brutal. Arizona alleged 1.5 million blood tests sold to its residents on misrepresented terms. Holmes got 11 years and 3 months. Balwani got 12 years and 11 months. Together they were ordered to pay $452 million in restitution to twelve victims.

Here is what makes this story unfinished. Alan Eisenman, an early investor owed almost $150,000, says he has not received a penny. In March, a judge cut Holmes’s sentence from 135 months to 123. She is scheduled to move to a halfway house near Austin in August 2027, with release set for February 2030. Her commutation request is still pending. And prosecutors told the court she has told national media she plans to return to health technology, while advising her partner’s startup, a company the government says resembles Theranos. The machine that built her is still running. The hype, the boards, the covers, the fear of missing the next big thing. Theranos did not break Silicon Valley’s incentives. It just showed us exactly how they work.