/cdn.vox-cdn.com/uploads/chorus_image/image/59937901/956956808.jpg.0.jpg)
On this episode of Recode Decode, hosted by Kara Swisher, Wall Street Journal investigative reporter John Carreyrou talks about his new book, “Bad Blood: Secrets and Lies in a Silicon Valley Startup.” Carreyrou explains how Stanford dropout Elizabeth Holmes’s company raised nearly $1 billion for blood-testing products that sounded too good to be true — and they were.
You can read a write-up of the interview here or listen to the whole thing in the audio player above. Below, we’ve also provided a lightly edited complete transcript of their conversation.
If you like this, be sure to subscribe to Recode Decode on Apple Podcasts, Spotify, Pocket Casts, Overcast or wherever you listen to podcasts.
Kara Swisher: Hi. I’m Kara Swisher, executive editor of Recode. You may know me as a fan of old vampire movies; my favorite is “Theranosferatu.” (Eric, you’re terrible.) In my spare time, I talk tech, and you’re listening to Recode Decode from the Vox Media podcast network.
Today in the red chair is John Carreyrou, an outstanding investigative reporter for the Wall Street Journal. I was just complimenting him effusively. He’s the author of a new book called “Bad Blood: Secrets and Lies in a Silicon Valley Startup.” It’s all about his fantastic reporting on Theranos, a blood-testing startup that sounded too good to be true, and it was. John, welcome to Recode Decode.
John Carreyrou: Thank you for having me.
I love the name. Of course, you had to call it “Bad Blood,” right? You had no choice?
Yeah, I came up with that title early on. That was actually the title of the book proposal.
Yeah, so was there any other blood things? There will be, there wasn’t?
I didn’t really think of any. That was the one that was screaming at me from the beginning.
Right. Yeah, absolutely. It’s the perfect one, and it’s a great tale, and I read all of your stuff in the Journal. I want to talk about how you did this. Let’s begin by talking about your background really briefly. People like to know where people come from. Give me the quick John file.
Sure. I started out in journalism in the mid ’90s working for the sister wire service of the Wall Street Journal, which is called Dow Jones Newswires. Then I joined the Journal in its Brussels bureau back in 1999. I’ve been at the Journal ever since. I’ve been based in Brussels, Paris and New York for the past 12 or so years.
Was health care your area for the whole time? No, I don’t think so.
Health care became my specialty when I moved back to New York. I came back and joined the Health and Science Bureau and covered the pharmaceutical industry for a year or two, and then joined the investigative team, and carried that expertise over to my ...
Why that area?
It’s what appealed to me the most coming back from Europe as a foreign correspondent. I knew I’d have to specialize. When you’re abroad, you get to cover everything. When you’re back in the states, you obviously have a narrower slice of real estate that you’re given, and so I wanted that slice to be interesting and really engaging to me, and health care was that. The Journal, as you know, had a fantastic history of great investigative and explanatory reporting in medicine and science. I was interested for that reason.
You did that for many years. How did you get onto this story? Because one of the things I was talking about earlier is, it was astonishing, there was a series of stories by tech reporters about Theranos, and I was thinking we never wrote about it because we thought it was a health care startup, and we didn’t have someone that could really do a great job at it. That’s not an excuse, it just wasn’t a tech company to us. But no health care reporters looked at it very carefully either. It was betwixt and between as a company.
So talk about what attracted you to it. Explain what Theranos is, for those who don’t know. They should know, of course, it’s a well-known disaster.
Theranos is a Silicon Valley startup that was launched by Elizabeth Holmes when she actually dropped out of Stanford when she was 19 years old in 2003. Her vision was to build a portable blood-testing device that would run the full range of lab tests from just a drop or two of blood pricked from the finger. That was the original conceit of Theranos.
Right, which is the great dream of many people. I have a blood issue, and so I get blood tests a lot. They’re always dragging a lot of blood out of you. It would be the great dream of anybody that had it.
Right, and this is something ...
And money and stuff like saving money.
Right, and it’s something that researchers have been working on for 20 years now, both in industry and universities. It’s a nut that hasn’t really been cracked. People have been able to come up with tests that it can do three or four off a drop of blood, but not 50, and much less hundreds.
These panels. They’re called panels, right?
Right, these huge panels. She was saying that pretty much every test that they did in big labs, in big reference laboratories, that she could do with her technology off just a drop or two of blood.
Talk about how she got that idea? Because you’re saying it’s a startup in Silicon Valley, but it just happened to be a startup that is in Silicon Valley, not that is a Silicon Valley startup. I know that’s parsing words, but she attracted a patina around her that it was a tech company.
Right. She very much positioned Theranos as a tech company, even though, as I argue in the epilogue of the book, it was actually more first and foremost a health care company and a medical company, which she very much ...
Medical device company.
Medical device company.
Which there are many more in Silicon Valley. We’ll get into that later.
Right, medical technology company. She very much wanted to drape herself in the mystique of Silicon Valley, which most people would understand as what came out of the chip industry and what came out of the computer industry, and what is now ...
Or the internet industry.
The internet industry now. She headquartered her company at first in East Palo Alto, or Menlo Park on the edge of East Palo Alto. Then in Palo Alto proper, just across the street, essentially, from the Stanford campus, and that was part of her desire to really channel the image and the trappings of Silicon Valley and make Theranos in the public eye a tech company, when in fact it was a medical company.
Yeah, that was interesting, and also locating across from Stanford Hospital, the medical school, which has got an enormous reputation. Talk about how she did this. Then I want to understand how you got to her.
All right. When she first dropped out, she actually had this vision for what she called a Thera-patch, which was like an armband that would have microneedles, and the microneedles would draw your blood and diagnose whatever ailed you.
Do microneedles exist? No?
At the time, microfluidics were a couple years old at that point. There was a Swiss scientist who had figured out that you could repurpose that micro-fabrication techniques used to make computer chips and use them to make tiny channels that would move tiny quantities of liquid. That was sort of like when ...
It certainly sounds good, right?
Yeah. Things in that space were certainly hot, and there was a lot of research going on. She sort of surfed on that, and she quickly abandoned the Thera-patch, the armband, because that was really too futuristic.
Hard.
That was basically science fiction.
Yeah. I’d like an invisibility cloak, John, but not happening today.
Right. The first pivot was to something that seemed a little bit more feasible, which was a toaster-sized microfluidic device in which you would slot a cartridge, and the cartridge would have your drop of blood. For a couple years, they actually tried to work on a microfluidic device, and they couldn’t get anywhere with it.
Her background was not medical, correct?
Her background was that she had dropped out of college after barely a year and a half of undergraduate studies. I don’t think you could say that she had any expertise whatsoever.
No. I’m making that point.
Right, right.
I’m making that salient point.
Yep. After a couple of years of trying the microfluidics, they abandoned it, and they pivoted to what was essentially a converted glue-dispensing robot. One of the engineers at Theranos ordered a glue-dispensing robot from a company in New Jersey called Fisnar, and then reprogrammed it, put a pet at the end of the robotic arm, and programmed the robotic arm to sort of mimic the steps that a lab scientist would do on the bench to test blood.
So it would be a robot blood tester.
Right, and it was pretty rudimentary. This robotic ...
None of these are bad ideas, but go ahead.
Yeah, but that was a big step down from microfluidics.
Yeah, I know.
Anyways, she had this black-and-white case custom designed by Yves Behar, the Swiss industrial designer.
Yes, I know what you ... You don’t have to say it like that. You can say Yves Behar. It’s fine, but I like the way you said it.
I’m half French. I grew up in Paris, so I tend to pronounce ...
He’s fantastic as he sounds, but go ahead.
Yves made this sleek black and white case with a diagonal cut.
Which is his thing.
Right, that hid the robotic arm, but it couldn’t hide the loud grinding noises that the arm and the pet made, but that was essentially the first device — or rather the second device — and it was called the Edison. She called it Edison after, of course, Thomas Edison.
Yeah, yeah.
Then the third iteration of that technology, which they started working on in late 2010 is what she now calls the Mini Lab. The Edison could only do one class of test known as immunoassays, which are tests that use antibodies to create chemical reactions. The Mini Lab was supposed to do more than just immunoassays. It was supposed to do general chemistry assays and blood cell counts, etc.
It needed to include in a small space many more lab instruments. It ended up being taller and deeper and weighing a lot more. We can get into this, but essentially by that time — she went live with the blood tests in Walgreens stores in the fall of 2013 — the Mini Lab, the last iteration of the device, was a totally malfunctioning prototype. She dusted off the Edison, which could only do a few immunoassays, and for the rest of the 200-plus tests that they advertised on the menu, they used commercial analyzers.
Meaning they sent it to the lab.
No, what they actually did is they bought Siemens’ machines and they hacked the Siemens’ machines to adapt them to small fingerstick blood samples. The main two things they did was they created these cups to go under the Siemens’ machines that were half the size of the regular Siemens’ cups.
Of blood?
Of blood. Yeah. Then the diluted the blood. Those two moves dealt with a problem called dead volume, and essentially making the cups smaller made the needle that went in the blood and aspirated the blood closer to the bottom, and diluting the blood created more volume.
But this came with problems, which is that the tests become less accurate, because you’re pre-diluting the blood before you put it in the Siemens’ machine. Then it gets diluted again as part of the Siemens’ machine’s testing protocol. You’re tampering with the device more than you should be.
Right, in order to get the results.
Right, and you’re also diluting the concentration of the analytes you’re trying to measure to levels that are beneath the range that the FDA has approved for the machine.
Sure. So obviously the big question is, oh, what a lot of fraud. Why?
Right. As I explained in the book, she approached Walgreens and Safeway in early 2010 claiming that she ...
Mm-hmm, seven years after she started.
Right.
Right.
Claiming she had a device that could do all these tests.
Right. How many total did she say she could do?
She said around 200 at the time, early 2010.
Okay, which is fantastic. If anyone’s seen a panel thing when you have a blood test, it looks like there’s about 200 on those charts.
Right. The problem was that was a lie. That was a big lie.
Yeah, of course. Yeah.
The Edison could only do, even if you assumed that the Edison could do many immunoassays, the Edison could have theoretically maybe done 50 immunoassays. The reality is that none of those immunoassays had been validated. The Edison could certainly not do these other classes of tests, which she was representing to these retail partners that it could do.
Give me an idea of what tests. Give a regular ... Again, I know about this, because I get my blood drawn a lot, and it’s a lot of blood. It’s constant. It’s a huge amount.
An example of an immunoassay would be a vitamin D test or a PSA test, which is the prostate cancer test, essentially. Then if you’re measuring cholesterol, that’s a general chemistry test.
Which is a common one.
Which is a common one, right.
Thyroid.
Electrolytes. No, thyroid, a lot of those are immunoassays, but if you’re looking at very common tests, like cholesterol or electrolytes, potassium, etc. ...
Potassium.
Those are all general chemistry tests. Those are tests that the Edison machine could not do.
To what end do you do these tests? I’m sorry to make you dumb it down, but for people ... It’s in order to find out various diseases and issues, because blood, of course, has so much information about it.
Right. There’s a lot of information. Some of these various tests would allow you to gauge whether your organs are functioning properly and are healthy, like your liver. Cholesterol, of course, has to do with whether you’re in danger of having heart disease. The bottom line is that these are among the most prescribed tests. If you get an order from your doctor, from your general practitioner even, he’s very likely to have you tested for what are called general chemistry assays, which the Edison machine could not do.
Could not.
She went to these two companies in early 2010, asserting that her technology could do things it couldn’t do, and she actually only started working on the next iteration of the technology at the end of that year. By the time she went live two or three years later, the latest iteration of the technology was a prototype that didn’t work.
Yeah, so she went to Safeway and Walgreens, right? Was it Walgreens?
That’s right.
Which are eager to be getting into this space. Let’s set the business. A lot of these companies in retail are trying to compete in lots of different ways, and one of them is to create these mini clinics. There’s all kinds of activity around that. It’s in order to draw people in and to cut costs. There’s a whole lot of this also going on in Silicon Valley with Forward and One Medical and everything else, this idea of simplifying the common medical procedures, essentially.
Yeah. It wasn’t just about bringing in more foot traffic, but they also hoped to share in the extra revenues and profits from the Theranos blood tests. Safeway, for instance, went and spent $350 million to create these beautiful clinics within its supermarkets and about half of its store. You’re talking 800 or 900 stores around the country where they did these renovations. Safeway’s then president, Steve Burd, thought it would easily pay for itself, that the investment would easily pay for itself, because their aggressive assumption was that they would create $200 million in new revenue per year.
In blood tests, or whatever. They do other things, obviously. They do vaccines and things like that.
Yeah, so those revenues never materialized and the $350 million that Safeway invested in renovating its stores just went down the drain.
Yeah, yeah. Then Walgreens bought Mini Clinic? I can’t remember all the different permutations of these, but pretty much all of them got into it. It’s a natural adjacent business for Walgreens or a CVS or something like that.
It makes sense if the product is validated and respects regulations. I think it does make sense.
Why wasn’t it validated by those two? I’m going to get to the VCs later and the rich people of Silicon Valley, but these are two companies that ... They’re relying on this as an object. When they buy a blood pressure cuff that they have there in the Safeway, it’s assuming that you get the right blood pressure from it. Usually those are just toys, I think, in those stores. But talk a little bit about that, why they didn’t.
Yeah, this is one of the parts of the story that’s really almost incomprehensible, is Walgreens’, in particular, lack of due diligence and lack of verification. Theranos again and again over the course of three or four years was able to keep Walgreens at arm’s length.
At one point, as I describe in one chapter of the book, Walgreens hired a full-time lab consultant named Kevin Hunter, and his role was going to be to help Walgreens vet the technology. Elizabeth and her boyfriend, who’s the No. 2 at the company, Sunny managed to get Kevin Hunter, after a while, excluded from their weekly video calls and from their in-person meetings, because he was asking too many questions that they weren’t happy with.
How did she do that? How did they do that? And why did Walgreens accept this?
Well, Walgreens, as my sources explained to me in the course of reporting the book, Walgreens was terrified that the Theranos technology would end up in the hands of a rival, and in particular in the hands of CVS, its larger rival based in Rhode Island. There was this fear of missing out that explains a lot of Walgreens’ sort of credulity and lack of due diligence.
Yeah, I could sort of expect it from a Safeway, but from a Walgreens whose business it is ...
Right. The dynamic on Safeway’s side was that Elizabeth had this relationship, this close relationship with Steve Burd, who was the CEO, and basically she only talked to him. They had a war room in Pleasanton, California, which is Safeway’s headquarters location, and they would discuss the Theranos partnership once a week. Steve Burd would always participate either in person or by phone. Whenever issues came up and Safeway executives wanted to ask Theranos questions ...
Tougher questions.
Steve Burd would say, “I’ll take care of it. I’ll take this to Elizabeth.” That continued until Steve Burd finally stepped down from the CEO-ship of Safeway midway through 2013.
Which is interesting, because when a CEO intervenes ... nobody speaks up when a CEO intervenes in those personal relationships.
Right, and by the middle of 2013 when he finally stepped down, there were a lot of qualms among Safeway executives. There were a lot of people who were frustrated and angry with Steve Burd for essentially giving Elizabeth Holmes this cover. At that point, she had been dangling before them this partnership for three years, and it was still not happening, although she did get the Theranos blood test used in the campus clinic in Pleasanton at Safeway.
But deployed to any of the stores?
It was never deployed to any Safeway stores.
Thank goodness.
I think that has to do with the fact that Burd finally stepped down in mid 2013, and then Elizabeth refused to speak to anyone at Safeway. Then Safeway executives, after Burd left, had to deal with Sunny, her boyfriend who was the No. 2 of the company, or with her younger brother and his fraternity friends. From there, the relationship kind of gradually fell apart.
They paid Theranos for this, or not?
They lent Theranos $30 million on top of the $350 million they invested in redoing their stores.
Right, right. On top of that. Then Theranos already had its own money, too. Why play those games with these companies if they’re not going to deploy them?
Well, this is essentially part of the fraud. What happened is by late 2010, Theranos was running low on money, needed to raise new money and was planning for a big fundraising round. It went live in Walgreens stores in September of 2013.
How many?
At first it was just one in Palo Alto, and then it was a second one, and then it grew to about 40 or 45 in the Phoenix area. Then it could suddenly go to prospective investors and say, “See?”
“See, it’s working. It’s in the field.”
“It’s working. We’re live. We’re commercial. It’s not even that we’re promising you that we’re going to be commercial soon. We’ve already done it.” A lot of investors were sold by that, which was that you could already go to Walgreens stores and get the service, get the product.
Right. Absolutely. When you think about how that ... It’s not the Ponzi scheme of it. It’s something else. It’s that it’s a very Silicon Valley thing. If that guy’s in, then that guy’s in and that guy’s in. That’s how you judge things. What was interesting — and we’re going to get to when we get back — is how many Silicon Valley, mostly guys, were in and it caused everyone else to get in. If a smart guy like Larry Ellison was in this, it’d have to be a sure thing kind of thing.
Right, although I would say in Ellison’s defense, and even in Draper’s defense, although he’s been attacking me, I would say ...
Oh, he does that to everybody.
I would say in their defense, they came in early. Draper gave her her first million dollars, and Larry Ellison came in the second round in early 2006. Ellison invested when it really was ...
An interesting idea. I could see it.
It was an interesting idea.
I could see him being very interested in it.
He decided to make a bet on it the way people in the Valley make bets on dropouts who have an unproven concept.
Absolutely.
Some of them, they know some of them are going to work out. Many of them are not. Most of them are not. I wouldn’t say that Draper and Ellison were defrauded the way later investors were defrauded.
Exactly. All right. When we get back, we’re going to talk about that, because the making of this into a tech company is really an interesting part of this and I think was a part of the way that pushed it forward in a lot of ways. We’re here with John Carreyrou. He’s the author of “Bad Blood: Secrets and Lies in a Silicon Valley Startup.” We’ll be back in a minute after we take a quick break for a word from our sponsors.
[ad]
We’re back with John Carreyrou, the author of “Bad Blood: Secrets and Lies in a Silicon Valley Startup.” It’s about Theranos, obviously. Blood, Elizabeth Holmes and more.
Talk about the Silicon Valley part of this. Located in Palo Alto, I saw her a lot around Silicon Valley. As I told John, she was putting it off as a tech company. We didn’t cover it at Recode, but it was very much covered by tech reporters more than health care reporters, which I found really interesting at the time. She had quite a presence. She had quite a ... She’s a striking person, interesting. She’s one of the few women CEOs. She had a lot of affects, like the turtleneck and the dramatic profile pictures and things like that. It was an unusual situation, and I think people try to put more women on covers and things like that. Talk about that, why she portrayed it that way.
Right, so her idol was Steve Jobs.
Yes, indeed.
She was absolutely obsessed with Steve Jobs and with Apple. Early on, as early as 2005, 2006, when they were still working on the microfluidic version of the technology, she would call it the iPod of health care. It was really conflating what she was doing, which was she was trying to do a medical product, with the tech industry and with Steve Jobs. Steve Jobs is the computer industry. It’s the internet.
That’s a very persuasive thing, because it’s instant thing to think about. It doesn’t dumb it down. It actually is a great market. It gives you a good visual of what she’s doing.
Right, but it’s the wrong role model if you’re trying to create a medical product.
Of course.
As you know in the San Francisco Bay Area, there’s this other part, you could call it a valley which is south San Francisco, which has all the biotech cluster.
They do, yeah.
There’s real science going on in South San Francisco and in some of the towns.
Berkeley.
Yeah, to the north and to the south. There’s a lot of medical diagnostics as well.
Genentech, others. There’s so many.
Right, so she could have chosen any one of those companies or founders or CEOs as a role model, but she chose a founder who had done his pioneering work in the computer industry, not in medicine. I think that was an enormous flaw from the beginning.
Why was that? Why did she do that? I could make the joke I knew Steve Jobs. Steve Jobs was not my friend, but she was no Steve Jobs. Why did she do that?
I think she was just taken in by the razzle-dazzle of the computer industry and of the internet industry, and also she was, I think, taken with the money, the valuations that some of these companies could reach. Facebook, Mark Zuckerberg was coming up around the same time that she was, and I think she was very aware of his rise, and I think she wanted to replicate it.
Also the money, the money. One of the jokes is there’s not enough rat holes to shove it all down, and there’s a lot of rat holes. What did she do? Talk about this, because I think there’s two things. There’s prominent investors, like George Schultz and several others. You could talk about them, and then the tech investors, who were Larry Ellison, Tim Draper and others.
Right, and again, those investors — Ellison, Draper — came in early. I would not say that they were defrauded. The investors who came later, and especially those who came in 2013 and after, because when you look at Theranos, the bulk of the money, all together she raised about $900 million, and about $650 million of that was raised after 2013. That’s when she used the fact that they had gone live in Walgreens stores as sort of the validation that Theranos ...
Who was the biggest investor in that?
The biggest as a family were the Waltons, the heirs of Sam Walton.
Walmart.
Who founded Walmart. They put in $150 million through to companies. The biggest single investor was actually Rupert Murdoch, who put in $250 million in early 2015.
Personal money.
Right. Betsy DeVos, the education secretary, her family put in $100 million.
Yeah, they’re trying to make it back now working for the ... Sorry. You don’t have to say anything.
The Coxs of Atlanta, Cox Enterprises put in another $100 million. Carlos Slim, the Mexican billionaire, put in 30 million. The Oppenheimer Dynasty of South Africa used to control De Beers Diamond Company, they put in I believe 20 or 25 million. A Greek shipping heir put in another 20 or 25 million.
Stavros?
One of his heirs did.
Yeah.
What you’ll notice among this later group of investors.
Rich people.
None of them are VCs with any experience in health care, medical technology.
I was thinking dumb rich people, but go ahead.
Right. These were essentially family officers, very rich families, billionaire families that were not sophisticated investors. I happen to know that some sophisticated investors, like Sequoia, actually knocked on the door of Theranos around the 2013, 2014 period.
Why not? Yeah.
They’d caught sight of Elizabeth, started to see that her profile was rising, and they were just curious. She said she wasn’t even interested in talking to them, because she didn’t need the money, because she was getting plenty of money from investors who were not sophisticated and who would not ask the tough questions that she didn’t want to respond to.
They certainly would have, certainly would have. The original Silicon Valley ones, Larry Ellison, Draper Fisher Jurvetson, who else in there?
There was another Silicon Valley firm called ATA Ventures, which I believe makes a specialty of investing at a very early stage. They accompanied Theranos through several rounds. Again, they invested very early on.
It was just on a whim. There’s a lot of those happening, and especially in medical areas. There’s a ton of them now, Cardia, there’s Color, there’s 23andMe. They’re all looked at like that, I would say. Like, “Okay, this is interesting. Some of it’s very promising. Let’s go into this.”
Right.
Talk about their other advisers. She created a whole stable of ...
The board.
The board.
The main way that she got away with what she got away with is that she surrounded herself with larger-than-life figures. Early on, she met Don Lucas, Donald L. Lucas, who was a very well-known venture capitalist and who groomed Larry Ellison and helped him take Oracle public in the mid ’80s. He became chair of her board, I believe, in 2006 after he invested. Really Lucas in the early years gave her credibility.
Cover.
Then Lucas, unfortunately, got Alzheimer’s disease, and so she turned in 2011 to George Schultz, whom she met through someone at Stanford. She managed to wow George Schultz, who I happened to know through my reporting is passionate about science. Pretty soon, they were meeting on a weekly basis and he was joining the board. Then she used her budding relationship with Schultz to meet all these other very famous ex-statesmen and military commanders who were all senior fellows at the Hoover Institution.
Just to be clear, it’s a Stanford Institution, it’s conservative, actually. It’s a more conservative bent on this campus of Stanford.
Right. It’s a think tank that has its building squarely in the middle of the Stanford campus. Among the other senior fellows there were people like Kissinger and Sam Nunn and Admiral Roughead, Perry, the former secretary of defense under Clinton, William Perry.
Yeah, it’s quite a place. I had lunch there recently, and I felt like I was going to get old white men poisoning, but that was just me. It’s a real place. It’s a really ...
One after the other, she recruited these guys who, by the way, were all in their 70s, 80s and 90s.
Yeah, they wanted relevance.
And who didn’t have any experience to speak of, except Bill Frist was once upon a time a surgeon.
This was the one. Yeah.
He was arguably one of the only two on the Theranos board who had anything to do with health care. She promised them stock in exchange for joining her board.
They wanted to get in on the Silicon Valley thing, too.
I think so.
Yeah, they couldn’t get into Facebook, but, “Hey, we’re still relevant. We’re going to play.” It’s hard. It’s very heady when you’re there and you’re not part of it, and you definitely ... Right now with bitcoin and crypto, everyone, I literally have billionaires being like, “I got to get some,” and I’m like, “Okay. Don’t you feel like you have ...” It’s fascinating to watch the FOMO mentality of everybody there, if you’re not onto the thing. I can’t imagine what it was like not to be part of the internet revolution.
Kissinger thought he was going to ... He had the shares in a trust for his children, and he told people that ...
(In a Kissinger voice) “I am going to have someday be rich.” He should have bought bitcoin.
His heirs were going to get really rich through Theranos. Undoubtedly, there was this quid pro quo going on. They thought they were going to get rich, and she used them for their reputations. These guys had sterling, all of them, sterling reputations. She also had Mattis, who had just ...
General Mattis?
General Mattis, who’s now our secretary of defense and who had just stepped down from Sencom at the time. He was another great get for the board. One of the investors that put in 100 million was this hedge fund, partner fund management based in San Francisco. When their guys went to Theranos, and they saw the board, that was almost enough to sell it. How could anything go wrong at a company with such a sterling board with, by the way, a lawyer who sat in on all the board meetings whose name was David Boies?
Right. To add to it, let’s bring him in.
How could anything go wrong?
Oh man, it did. Talk to me. Let me talk about how you got interested. I do want you to talk about George Schultz’s grandson, because I think that was your finest story, getting that turnabout was epic in so many ways. I hope it’s part of the movie, because that, to me, was the most astonishing moment in your story. There’s a time when you read a story and you go, “Holy Jesus.” You don’t get many holy Jesuses in your life with a story. Just recently, the Ronan Farrow one on Eric Schneiderman. There was a holy Jesus moment, but that one was that for me. Talk a little about how you got in. Why did you suddenly go, “This is funny?”
Well, I read ... The New Yorker published a story in mid December 2014, and it was another one of these profiles of Elizabeth, mostly fawning. I read it on the subway home, and I thought there were a couple things that were strange in there, including a quote she gave about how her technology worked, which sounded more like a high school student than a sophisticated laboratory scientist, and also the fact nothing that she had supposedly discovered had undergone peer review.
Right.
To be fair and honest, I would probably not have done anything, if it weren’t for the fact that three or four weeks later, I was approached by a tipster, a source of mine who had helped me with an investigative project I’d done the year before. He was a practicing pathologist in Columbia, Missouri. He wrote this relatively obscure blog, and he’d just written a short skeptical item on his blog after reading the New Yorker story, in which he expressed skepticism.
“This is bullshit.”
Right. He said it more politely and elegantly than that, but yeah, he was skeptical. After he wrote that, he was contacted by this group of people who had been these quiet skeptics and had some information indirectly. Then he got in touch with me. One of those people had recently been in touch with a Theranos employee who had just left.
Those are the best.
When the pathologist in Missouri got in touch with me, I had to pull on the string. It was essentially a third-hand tip. I had to get to the employee, because everything else was sort of conjecture and speculation.
That’s the key part. I try to do exit interviews on everyone who leaves companies. That’s how I get a lot of scoops, because you do.
Right.
You do.
Right. No, that was the key, is getting to that former employee who had just left.
That employee had information you wanted, obviously.
Yes. Well, first of all, he was scared to death. It took me a while to get him comfortable. I promised him confidentiality. He finally opened up to me. Then it turns out that because of the position that he was in, he was a high-ranking employee at the company directing ... He was essentially director of the lab. For that reason, he had tons of information. He was an unimpeachable source, but he was an anonymous source. I wasn’t going to be able to do the story with just one source.
From there, it became a game of corroboration. I tried to get what he was telling me corroborated by other ex-employees.
Essentially, if I could boil down your reporting: “This doesn’t work. This doesn’t work.”
Yes. He told me that they have this one machine called the Edison, and it doesn’t work. It’s only doing this handful of tests. Then they’ve hacked these commercial machines made by Siemens, and they’re doing most of the other tests with those.
Bad tests.
Those are problematic as well. Then they’ve got this other machine they’re working on in the background that he knew as the 4S. It was the Steve Jobs-inspired code name, because when they had started working on the Mini Lab in I think it was 2011 when the iPhone 4S had come out, she had code named it the 4S. That’s how obsessed she was with Jobs.
Yeah. No, I read that. That was a holy Jesus.
Right. He knew the Mini Lab was the 4S, and all he knew about it is that he knew that the people, his colleagues who had been working on it, told him that it was a box that didn’t work. What they were working on in the lab was the previous iteration of the technology, the Edison. That was limited and it didn’t work either. Then the hacked Siemens machines, the results from those were problematic. Yeah, from that point on, I knew this was potentially a big story.
Right.
Because it meant that not only had ...
She raised a billion dollars.
Right, and then potentially defrauded investors, but she was potentially putting patients in harm’s way.
Right. That was the key part of it. As you said in this thing, you used the term, in your piece, vaporware. The term vaporware was coined in the early 1980s to describe new computer software or hardware that was announced with great fanfare, only to take years to materialize, if it did at all. Who cares, right? Who cares if you don’t get another photo app or whatever?
You might be disappointed, but you’re not going to ...
Yeah, not particularly.
Your health isn’t going to be ... You might not even remember that the promise was made two years before.
Or something. There’s a lot of companies, there was General Magic, which eventually led to the iPhone. It really did. There’s a great new movie out about it. Some things just don’t work, but they’re great ideas, and they’re conceptually right, but either the timing, because this technology is not there, or the mobility is not there, or the network is not there, but the conceptual ideas are great. But this was different, because it was people’s lives at stake.
Right, and it’s an enormous difference. I think she had to have been aware that going live with blood tests that she hadn’t validated, that were not really reproducible, according to the company’s own data, she had to know that they were putting patients in harm’s way. That’s just beyond the pale and just hard to get your mind around.
We’re going to get to why she did that next, but just tell us the story about George Schultz’s grandson, because the grandson was working there. This is another thing. They put their various and sundry relatives in some of these companies. It happens all the time in Silicon Valley.
Right, so George Schultz was once secretary of state for Reagan. Before that, held two cabinet positions under the Nixon administrations.
Big executive, too.
He remains a revered figure in Republican circles, lives out there. His house is essentially a block or two away from Stanford.
Mm-hmm.
His grandson, Tyler Schultz, was graduating from Stanford in May of 2013 and then went to work for the company the following fall. After a few months, started realizing that he was uncomfortable with a lot of the practices there, and tried ...
He was a biology major?
He was a biology major, yep, and he tried to raise his concerns with Elizabeth and eventually was forced to leave the company on not good terms after eight months.
Because he was saying this doesn’t work.
Right.
This is a kid.
He tried to get his grandfather’s attention and tried to explain to him what he had seen and that he wasn’t the only one who had these qualms. His grandfather wouldn’t listen to him, so he ended up leaving the company and was quiet for about a year, until I started poking around after I made contact with the ex-laboratory director. Tyler heard that I was poking around through the grapevine and looked up my profile on LinkedIn. You know how on LinkedIn, you see when people look up your profile. I saw that he looked me up, and my first source had mentioned his name. I had already written him down as another potential source in my notebook. I immediately wrote him an in-mail message via LinkedIn.
Thank you, [LinkedIn CEO] Jeff Weiner. But go ahead.
I didn’t hear back from him for about a month. Then suddenly I was at my desk in a newsroom of the Journal in Midtown Manhattan, and my phone rings in the middle of the day, and it’s Tyler. He was very concerned and nervous, but I could also feel he wanted to speak. He’d been carrying this weight with him for a year, and here was a reporter who’s finally looking into these issues.
You talk about that weight, because it was agonizing. Your interview with him was agonizing, and his relationship with his grandfather was agonizing.
Right, although the scariest part only came later. I actually went ... A few weeks after Tyler and I made contact on the phone and he began telling me about what he knew and corroborating what my first source had said, I went to Silicon Valley and he and I had a beer at a beer garden in Mountain View. Then I came back to New York, and the company figured out that Tyler was one of my sources and essentially ambushed him. Theranos lawyers — actually Boies Schiller lawyers, but they were acting on behalf of Elizabeth Holmes — ambushed Tyler at his grandfather George’s house one evening and tried to get him to admit that he was talking to the Wall Street Journal and tried to get him to ...
Renege.
Well, first admit that he was talking to the Journal, and then recant what he had told the Journal and then name the Journal’s other sources. This pressure campaign went on unbeknownst to me for months, because at that point Tyler broke off contact with me. He had been talking to me via a burner phone and through a sort of a Yahoo address that he had made up just for our communications. I didn’t have his real phone number. I didn’t have his real email address.
Once Theranos started putting the screws to him, I stopped hearing from him. I didn’t know for months what had happened to him. In fact, I only found out about a year later once my first story had been published, and then I did a bunch of follow-up stories. It had blown up into this big scandal. I flew back to Stanford and met Tyler in a classroom in the materials science building. He finally told me what had happened. I was stunned.
At that point, he was telling me, even though his lawyers had ordered him not to speak to me. He was telling me off the record, and I promised to keep it between us until he gave me the permission at some point in the future to write about it, which he eventually did.
Then a couple days later, I’m back in New York, and Tyler sends me an email.
The lawyers are back.
He tells me that Theranos has contacted his lawyers, and that they know that we’ve met again.
So they were following him.
They were following Tyler. They might have been following me, too.
Then of course they put the screws to his relationship with his grandfather, which I thought was the most sad.
Right, and so they became estranged. I think they’ve since been able to repair the relationship a little bit.
Still.
I think George Schultz has finally seen the light, but this has been an incredible ordeal for Tyler. It’s something that I don’t wish on any 25-year-old.
What’s he doing now?
He’s actually got a startup. It’s a diagnostics startup. I believe he’s in the midst of raising money for it.
Oh, good for him. All right, so when we get back, we’re going to talk more about sort of what’s going on in health care and what happened to Elizabeth Holmes — or what’s going to happen to her, really, because it’s ongoing. We’re here with John Carreyrou. He has a new book out about the Theranos scandal called “Bad Blood: Secrets and Lies in a Silicon Valley Startup.” We’ll get back, we’ll talk more.
[ad]
We’re here with John Carreyrou. He is a fantastic reporter for the Wall Street Journal. His book, “Bad Blood,” is about the Theranos scandal. We’ve been talking about how it unfolded. They came after you too, you’re talking about them coming after George Schultz’s grandson, who was on the board, for essentially giving you the truth, telling you the truth. They came after you before you were publishing this. Lots of threats, lots of big-name lawyers.
Right. The first two months, I would say, they gave me the runaround. Then one day I was in the newsroom, and she was on Charlie Rose.
Oh perfect.
It was like the latest in a series of media appearances, TV and magazines and so forth. It seemed I was the only reporter in America that she wouldn’t speak to. I called her PR handler, and I said, “You guys are not going to be able to ignore me forever. You need to engage with me.”
They finally agreed to set up a meeting, but it wasn’t going to be with Elizabeth Holmes, nor was it going to be with Sunny Balwani, her boyfriend and No. 2 at the company. It was going to be mostly with lawyers. At first, they said, “Come to the offices of Boies Schiller in Manhattan and we’ll talk there.” I said, “Yes.” And then I thought better of it.
No.
I thought, “No, that’s going to be like marching into the lion’s den.” So I said, “Why don’t you guys come to me?” They did in late June of 2015. I fully expected fireworks so I had Mike Siconolfi, the Journal’s investigative editor.
Great, a reporter too.
Who’s great, and Jay Conti, who’s our lawyer who worked with us on sensitive matters. They came with essentially mostly lawyers. There was David Boies flanked by two Boies Schiller partners. Heather King, who had just recently left Boies Schiller and become Theranos’ general counsel. Peter Fritsch, who co-founded with Glenn Simpson the opposition research for Fusion GPS, which has since become notorious for its role in the Trump Dossier. And this lone Theranos executive named Daniel Young. Of the delegation of six or seven people, there was one person from Theranos.
Why did all these people protect her? Because like you were saying, Tim Draper attacked you, others for a while were giving you an enormously hard time. Why did all these people protect her?
You’d have to ask David Boies exactly what was going through his head.
Well, he’s got judgment issues all up and down. Let’s move onto Harvey Weinstein. You know.
I don’t know. Part of me thinks that David believed that she had real technology. He had done patent work for several years in a patent case for her. He had been paid. He and his firm had been paid entirely in shares.
They had to ...
They’d been paid 322,000 Theranos shares, which I think comes out to about $5 million at the valuation they were giving the shares at, which was $15 a share. That made him the legal advocate of Theranos, but it also gave him a financial interest in Theranos’ future.
It makes him impossibly complicit, really.
I think it makes him conflicted, but I think he also believed in the company.
Yeah.
That meeting we had in late June was surreal. We went around in circles for five hours, because they kept telling us that we had misappropriated trade secrets.
Oh yeah, that one.
We had stolen trade secrets, and we needed to return them.
Oh God.
If we made any moves to publish them, they would sue us.
Right.
Besides that, that a lot of our information was wrong, and that our sources were disgruntled former employees and so on and so forth. It was one of the most surreal meetings that I’ve ever participated in. And from there, the campaign — this is sort of what I call Theranos’ scorched earth counterattack — kind of escalated.
Why? I’ve dealt with Uber, and they eventually ... That was pretty tough, but this was beyond that, you know what I mean?
Right. I think they must have thought that there was a reasonable chance that they could get the Journal to kill the story.
But not looking into whether she was lying or not. There was none of that.
By Boies Schiller?
Yeah.
I don’t know exactly what Boies Schiller did.
All of them. It’s like Schultz, all of them. What was it? Let me just say, I’ve met a lot of jerky people. She wasn’t. I’ve met her personally. She’s rather pleasant. I don’t see her as ... She doesn’t seem menacing or particularly scary, like Travis kind of feels a little scary when you deal with him, a little bit tough.
She does have this charisma, though. I saw her later, more than a year later at the American Association for Clinical Chemistry’s annual meeting.
Yeah, she shows up.
She showed up in Philadelphia. At that point, the story had blown up. She was still incredibly impressive.
I saw her. I saw her at an event, too. I couldn’t believe she was there.
Right. One of the sources for the book, an early Theranos employee, used to joke even back then, this is in 2006 to 2008, that she could sell ice cream to Eskimos, that she was that great a saleswoman. I think her idol was Steve Jobs, and she did have this reality distortion field. Maybe you and I wouldn’t have fallen for it, but a lot of people did along the way.
What was interesting ... Why do you think she did this? Tell us what happened, what’s going to happen to her, what has happened, and what happens to the company.
Why she did this? She said when she was 9 or 10 years old when she was asked what she wanted to do when she grew up, pretty much the question everyone is asked at one point or another when you were a child, her answer was, “I want to be a billionaire.” She had this voracious ambition. She wanted to be an entrepreneur billionaire. She wanted to be the female Steve Jobs. She wanted the fame. She wanted the money. And she would stop at nothing to get it. If there was collateral damage ...
She should have just joined Facebook, but go ahead. There’s plenty of places. She wanted her own. She could have done it.
She wanted to be the first female billionaire founder. She wanted to be that. Simply nothing was going to ... She was simply not going to let anything get in the way of that.
C’mon, John. What is at the heart of that? I think about all these people I deal with all the time, you know what I mean? I was joking the other day. I was very difficult with someone, and these are comparatively, because I think people’s lives were at risk, it was a minor take-down on my part. They said, “When are you going to stop?” I was like, “When you stay down. Stay down.” It was sort of interesting why she, one, doesn’t stay down, and two, did it.
Right. That’s some of the more interesting, if you’re trying to psychoanalyze her is ...
I’d like you to.
She was found out, essentially, by me and the Journal in late 2015. She has continued with the charade of supposedly developing this product and keeping this company alive. We’re now two-plus years into this and she’s a couple months from completely running out of money, and Fortress Investment Group is going to seize the assets and liquidate them.
How many people are still there? Who works there?
125 people were there probably as of last week, but most of them have been told they won’t have a job anymore, a paycheck on June 12th. At that point, the company goes down to about 20 employees. Then they’re going to be there turning off the lights.
The assets. What are the assets?
The assets were mainly the patents. One of the things that Theranos was good at was patenting a lot of things.
Processes, the processes.
Fortress saw that, and Fortress went and extended its loan, its sort of vulture loan, to Theranos last year, got the patents as collateral. When the company’s liquidated, the main value is the patents.
It could be something, right? Or not? It’s a process. They patented the process.
I think Fortress is mainly going to play patent troll with those patents.
Right, if they come up with a similar kind of concept or idea. Could it be valuable?
I think eventually there are going to be breakthroughs in diagnostics, including with small samples. I just don’t think it’s going to come out of any work that Theranos has done.
Will she ever be a billionaire, John?
I think she’s potentially facing indictments.
Yes. That was my point, yeah.
She and Sunny Balwani, her ex-boyfriend, the ex-No. 2 at the company, are potentially staring down indictments. If you read the SEC’s complaint from a couple months ago, the case for securities fraud ...
Southern District New Yorker who ...
No, it’s the U.S. attorney’s office in San Francisco that has been conducting a criminal investigation. My sources tell me that it’s very much proceeding apace. It looks very likely to result in indictments. I think that she’s looking at ...
But she has not admitted anything.
She has not. She didn’t admit ...
There’s got to be emails, or maybe not. She might not have emailed.
She didn’t admit wrongdoing as part of the SEC settlement under the Obama administration, if you remember. They used to require some sort of admission of wrongdoing. Then under the Trump administration, they’ve stopped doing that. They didn’t require her to do that.
Oh, she’s lucky Trump got elected, I guess. Wow.
I think criminal charges are very ... There’s a very distinct possibility of criminal charges here.
Is there an email trail of where she’s saying, “I know this isn’t true?” Probably not. She’s too clever for that.
I don’t know if there’s an email trail. I can’t say that I have email evidence myself, but when you read my book and you put it all together, it’s impossible to think ...
She didn’t know.
... that she wasn’t aware of what was going on.
That might be a defense she’ll have. I want to finish up. We have just a few more minutes to talk about this. Despite Theranos, which I think put a black eye, it was Theranos and then Uber. It was bad Silicon Valley all over the place. There are a lot of really interesting medical device companies, a lot of prominent tech people involved in medical. I was saying Color, which is a cancer genomics thing. There’s 23andMe, which has gone through a lot of problems with the FDA but seems on track to be creating a really interesting company. Maybe you’ll tell me I’m wrong, but it seems like all these areas, a lot of stuff around DNA, a lot of stuff I’m thinking — Cardia, which is interesting, the EKG stuff. Every day a new one.
Again, I’m not a doctor. I don’t have any expertise in the area. How do you assess that? Because tech is getting big. Larry Ellison, speaking of which, life extension, Google, life extension. Nobody wants to die, all kinds, AI, putting your brain on ... Everything.
All these companies are going to have to deal with peer review, and they’re going to have to deal with the FDA. There’s just no going around it. Elizabeth Holmes tried to get around it, and this is the result.
Well, Cardia just got an FDA approval on its technology. I’m saying they are cooperating. Where do you see tech playing in the health care space?
I think it’s only ...
Not biotech, but tech, AI and things.
Yeah. There is a convergence, and I think the convergence is going to continue, and it’s warranted. There are a lot of engineering advances that lend themselves to being married with medical advances. I don’t think that’s necessarily a bad thing, but I think the Theranos story is a cautionary tale about how not to go about it and how not to model yourself too much after the computer industry and the traditional Silicon Valley, because you have to remember, always remember, that patients’ lives and patients’ health is at stake.
Absolutely, but I don’t see them doing that. It’s really interesting, because right now, I just was at something the other day, and they’re like, “AI will replace radiologists.” When you think about it, probably that’s the case. Probably radiologists aren’t as accurate as a computer would be looking at screens or diagnostics. Eventually, AI and whatever it becomes will be good at diagnostics, will be good at basic health care.
I’ve seen robots that are surgical robots. You’ve seen some of these. I saw a thing, I think it was at MIT, that dealt with minor ... There’s 10 or 12 things that people come to the emergency room, they’re relatively easy and cost-saving if it wasn’t people dealing with them. They’re using robot faces to do it. They couldn’t get the eyes right. They can’t ever get the eyes right, but you see it. You see them moving in that direction. You see the money going towards it.
A lot of these tech people are getting older. They absolutely don’t want to die. They’re doing everything from nootropics to micro-dosing. I just interviewed Michael Pollan about the LSD. It’s a really interesting shift for Silicon Valley.
Right. I was just watching a season four episode of the HBO show the other day.
Yeah. Right. The Blood Boy.
The Blood Boy and the transfusions.
Yeah. They’re doing it. Trust me. It’s fascinating. Their interest in it is huge, and they feel like they can apply computer principles to it, but you’re saying maybe not.
Well, that may be, but they’re going to have to prove it by publishing in scientific papers and by subjecting their innovations to peer review. They’re going to have to, to some extent, to a large extent, accept the rules of medical research. Elizabeth Holmes and Theranos did not accept the rules of medical research and tried to get around them and to ignore them. The result is this fiasco.
All right, but in that case, the product doesn’t work. Do you see the medical industry being too slow and too not willing to break certain things? The famous Facebook motto is, “Move fast and break things.” That’s not working out well for them right now, but conceptually, on the flip side, what moves the medical industry forward? Because you can see our health care system is broken. The way we get diagnosed, there’s got to be something, or not, not at all. I’m just thinking as someone who just is relatively intelligent, but it seems as if it should be. The way medicine is done has no ... even just records, just everything, just doesn’t seem to get moving.
I think there’s been a lot of progress. There’s been innovation. I would argue that the U.S. health care system is broken for reasons other than innovation. Could some innovation happen even faster? Could it benefit from sort of the boldness of people in the Valley? Sure, but I still think that ultimately they will have to deal with publishing, with subjecting their innovation to peer review, and with explaining what they’re doing with some degree of transparency.
Right. Yeah. They don’t seem to like to do that. I was at an event the other day and they were all complaining about the FDA. I was like, “Too bad. You have to ...” They were saying, “What about this? Why can’t we do this? Why can’t we?” It’s a really interesting thing. Not everybody is a liar like this group at Theranos. They’re liars, if you really want to boil it down.
That’s right.
Not all of them are.
I like to think this was an outlier, that this was a lot of bad things that were extreme that aren’t going to be common going forward in Silicon Valley. In other ways, it was also representative of this certain hubris and arrogance, because she and her boyfriend Sunny didn’t even really make the effort to understand what was going on in this field of blood diagnostics and what other companies were working on.
It just looked like a good thing.
They just thought, at the same time as they were committing fraud, they also had convinced themselves that their malfunctioning prototype was the greatest thing humanity had every worked on.
Well, it’s interesting. It’s like any great grifter. They believe their own lies, right?
Right, right.
Seriously, when you think about it. But it’ll be interesting to see where health care is going with the combination of tech. Then lastly, when you think about what’s interesting now in health care, what are you fixated on now? Besides this is getting made into a movie, right?
Yeah. Yeah.
Jennifer Lawrence is attached, and apparently Deadline did announce the screenwriter.
Right. A screenwriter has been hired to do the screenplay. The movie will be directed by Adam McKay and Jennifer Lawrence is attached to play Elizabeth Holmes. What am I looking at?
Who’s playing you?
I don’t know yet.
You’re in it though, right?
Probably some unknown actor.
Are you in it? You’ve got to be in it.
I don’t know. I haven’t seen ... The screenplay is yet to be written.
You should be in it.
It’s going to be written.
You’ve got to be in it, come on.
I guess I hope I’m in it.
It’s got to have you, and then George Schultz.
And that some cool actor plays me.
Who’s playing George Schultz? Come on. Think about it. Who would you like to play you, John? Come on.
I think the guy who plays Thor would be the right choice.
Because he seems like a Wall Street Journal reporter. I’ve never seen a Wall Street Journal reporter that looks like Chris Hemsworth, I’m sorry to tell you, or else he’d be working there.
The diameter of our biceps is similar.
Yes, exactly. Yeah. What are you interested in next?
I haven’t thought all that much about it. The Theranos story was sort of an opportunistic dive into the field of blood diagnostics.
It’s still got some laps, too.
I hadn’t reported all that much about Silicon Valley, and so it also kind of plunged me into this world of Silicon Valley that I didn’t know. I would guess my next act is going to be in Silicon Valley, some other, not necessarily health care related.
Oh, please come.
Some other Silicon Valley saga.
We could use your help. We could use your help. Please come and ruin things. I’m leaving soon. You know what I mean? Someone’s got to go. Oh my God, it would be so good. It’s a good time.
I’m going to cover Silicon Valley from Brooklyn.
Okay. No. No. That’s what a lot of you do.
I’ll come often, though.
All right. Okay. I think you should, because there’s a lot going on there. There’s a lot of really intriguing, especially their not wanting to die or putting embeds in your eyes. They’re going places that should be interesting for someone like you to cover. I think I don’t know what they’re talking about, but you’ll be able to tell me.
Anyway, John, thank you so much. This is John Carreyrou. He is the author of “Bad Blood: Secrets and Lies in a Silicon Valley Startup.” It’s about the Theranos scandal, but he also writes about health care for the Wall Street Journal. It was great talking to you. Thanks for coming on the show.
Thanks very much for having me.
It’s a fantastic book. The stories were fantastic. You should read everything he writes.
Thanks very much.
This article originally appeared on Recode.net.