Hacker News Reader

Comment by neom | Hacker News Reader

I followed the case very closely, read the transcripts I could find daily etc. What Elizabeth did wasn't far from what many many many founders do. This was held to a legal standard, sure, but let's be real here... since when was the legal standard the SV standard? Is SV really the bastion of morality and legal ethics? Is that the new norm, good behaviour? Notably some investors refused to participate and iirc Draper basically said she flew too close to the sun, but it wasn't fraud. I'm sure it's going to be an unpopular opinion but personally if this is the new standard, the SEC should have a pretty long list of founders to try at this standard. The saddest thing of it all to me is that she was onto something, people are having success with her ideas today. She should have just not gone to market, she should have just quietly refined the tech behind the scenes, but as I said, she flew too close to the sun. Tesla and uber have literally killed people.

View on HN

> people are having success with her ideas today

Citation needed. The whole concept has a critical flaw, as pointed out by her engineering staff and medical advisors: blood tests require a certain minimum volume of blood to work, and that's inescapable because blood components are discrete, not continuous.

For example, a drop of blood is about .05mL, or about 1/20th of a milliliter. A common testing sample size in a normal lab is 3mL, or about 60x larger what Theranos claimed they could work with. Now, suppose you're testing for some cell that's may be present in concentrations of 10 per mL, or 1 per .1mL blood volume on average. In a conventional lab, then, you'd probably have about 30 of those in the 3mL sample. In a Theranos lab, even if it worked perfectly, you're equally likely to have either 0 or 1 present in the .05mL sample. In other words, you're 50% likely to get a false negative, even if the equipment is working perfectly, based purely on the tiny sample volume.

By analogy, if you poll 100,000,000 Americans, you might get a pretty good idea of an election's result. If you poll 3, you have no idea, even if the poll is perfectly designed. Well, that's what Theranos was up against.

NB: I'm not a lab tech, and the above is my understanding of it. That's my distillation of a lot of articles I read about Theranos, such as https://theconversation.com/the-rise-and-fall-of-theranos-so... .

No need to poll 100 million. 20,000 is pretty accurate.

OK, sure, but that doesn't change my point. You could run all of someone's blood through a scanner and then back into them to get a complete view, but a lab only samples a few mL of it.

Usually your sample size would be a good indicator of the accuracy of your poll/test/experiment.

Of course, like everything it could be biased by the used methodology.

https://www.washingtonpost.com/technology/2021/11/16/blood-s...

Maybe not quite as extravagant a claim as hers, hers was def a 100 year vision, but many people are working on and having progress in a very similar area. Heck even Tyler Shultz the kid who is on the road doing a press tour about how "he took down Theranos", is working on a Theranos-esq startup.

One thing that seems to be forgotten or lost in all of these armchair analyses is that while it is true that something like 3mL of blood is collected for a conventional lab test, the machine itself when it runs the test only takes a few microliters (maybe 10s of uL, all of this is depending on the test of course). The rest is frozen for follow-up or repeat testing or used for other tests or just discarded after some time.

EDIT: And my favorite part is the first step the Siemens machines they used for testing is to "dilute" the part of the sample it is going to use (addition of DI water, to get the blood to the desired concentration). That is hard-coded into the protocol, with a specific dilution amount, and it is done for all the official Siemens tests, as well, hah.

According to an article in Nature[1]:

> Holmes described the miniLab as “the most important thing humanity has ever built”. But at best, the lab could do immunoassays using microfluidics. The tiny blood sample had to be diluted extensively (for which there are no reference standards or precedents), leading to artefacts and spurious results.

That (and other reports) sound like they diluted way more than usual.

[1] https://www.nature.com/articles/d41586-018-05149-2

While it is published on their site, it appears to really just be a summary of the book by Carreyrou? And I have yet to see anyone actually publish any of the dilution ratios and compare that to any existing process, so it seems like an easy-but-easily-incorrect leap to conclude that it is "way more than usual".

While Theranos claims were absurd , it is not hard to believe there is scope for improvements in specific areas which is what OP is trying to say I guess.

1. The premise assumes there are not significant wastage in the current processes and those are near optimal. Any non optimal paths could potentially benefit from improvements

2. While there are some tests at that low concentrations not all tests have components at that low concentrations. Those could certainly be improved ?

3. Reuse and cross testing the same samples, or testing for multiple things at the same time etc, are all areas that could yield better outcomes than current standard.

4. Perhaps it is also possible that while primary marker for a test is not present in the sample, its presence in the blood could have left other markers that are detectable

I don't know anything about lab testing, if I can think of few approaches to solution for the low sample problem. Experts may have many other approaches so it shouldn't be surprising if there are improvements claimed ?

Many people are working very hard on solving these problems. Holmes simply claimed to have solved them all with one machine. They never had any idea of how to actually do that, never even tried. It wasn't like "If we use this new development in nanotechnology we might be able to...". It was just entirely made up, the whole time.

If I falsely claim to have invented a warp drive, I don't get credit for whatever work people do in that area in the future.

Not exactly how it works. There are lot of brilliant people working on lot of good stuff, very few get funding to do meaningful work. The job of someone like Holmes is to bring that funding.

Holmes was an agent to energize the industry in the sense that she was instrument to attracting a lot of new VC investors willing to throw money and experiment with new tech like this . I doubt she personally claimed to have invented anything new, even had Theranos succeeded her claim to fame would be only building the organization that enabled such innovation and not the innovation itself.

Even though Theranos failed as firm, a lot of smart people got funding to work on interesting projects while there and potentially either developed or started developing products because of holmes being able keep them and Theranos funded. A lot of new people are getting funding, because Theranos and Holmes opened this industry to VC funding when it was not traditionally so.

To your example, if you falsely claim to invent a warp drive, raise a lot of money, hire a lot of smart people who do interesting work while working for you, eventually fail as company and those people go on to build parts of the drive partly with experience they gained in your research. This kind of research was not well funded before you, Likely without you this would have never been a VC funded industry don't you have some hand in that ?

Do you think it's easier for people to raise money in the blood testing space now after this very high profile fraud? I've heard people say that it makes things much more difficult.

Also, Holmes's lies sucked up a ton of funding at the time. Not just the money given to her, but the investors who said "theranos already dominates blood testing, so we'll invest in another space"

This was a disaster all around for everyone but her.

>This was a disaster all around for everyone but her.

I don't know how you can say that, some of the trial evidence was heart breaking. Some of the emails, the texts back and forth between her and Sunny, what a mess, it sounded like she was totally losing her mind and stressed out to no end while being pushed and cajole by an unreasonably intense person behind the scenes. She was clearly devastated when the company collapsed, she very clearly believed in what she was doing, and now she is going to jail. That sound like beyond a disaster, it sounds like hell.

She clearly believed in what? The vision of low cost, multi function blood testing? Maybe she did, but she made almost no progress toward that goal. All while accepting credit for having already accomplished it. Not to mention that she's been living the lifestyle of a multi millionaire the whole time.

She's a human being and it is a sad situation. But this is justice.

Read the comments. Nobody cares how she feels now. Everybody cares only about their own opinion they have the need to express. For her it’s all devastating, for resonable people too, for majority of shitty people the main thing ia to grow in their own eyes while barking at someone else.

True, and there are chemistry tests that would probably be reasonably testable. For instance, blood glucose is already testable from a finger prick, so that’d be just fine. Still, a lot of people with the background to have informed opinions have spoken up to say that some of her claims weren’t possible on a theory basis, not just because of an engineering challenge.

This is a great breakdown and some great analogies.

What chemical would only be present in 10 per mL?

A mL of water has ~10^22 water molecules in it. Dilute substances are usually measured in ppm (parts per million). Even 1 ppm would have 10^16 molecules. I think your orders of magnitude are way off here.

> What Elizabeth did wasn't far from what many many many founders do

I'm not sure this is true. She told a lot of egregious lies to investors while trying to get money from them. She repeated those lies in the press and to anyone who questioned it. And I don't mean a lie like lying that your product is more effective than it is. I mean she told prospective investors about huge contracts with the defense department that did not exist. She told investors that her machines were being use in the battle field. These were lies that were never going to be true.

That's not what she said at all. At least not from the near daily coverage and transcripts I watched[1][2][3][4]

She said it was being trialed by the military, in reality what had happened was Mattis said he would be able to introduce the technology to the military. This type of "abstraction" is in very many pitches I've read over the years.

[1]https://www.youtube.com/c/LawyerYouKnow [2]https://www.youtube.com/channel/UCOShJ_GPEegXscrlKWfpwCA [3]https://www.youtube.com/channel/UCiWXvA_5bzvOFdbS353Ll-Q [4]https://www.youtube.com/channel/UCwCqasesYi60fSrOjlUSNzg

> She said it was being trialed by the military, in reality what had happened was Mattis said he would be able to introduce the technology to the military. This type of "abstraction" is in very many pitches I've read over the years.

That's not an "abstraction". It's a lie, and therefore fraudulent.

If you're seeing these sort of lies all over the place, you might be in a weird bubble. Startups are known for creative wordplay, but if you're straight up lying about something happening when it's not happening at all or even started, that's fraud.

Feel free to add me on linkedin and share your deal flow! I'd genuinely love to see the startups you see and are invested in. When I was still in startups, I encountered more than one experience where various investors begged me to lie to them. In one instance because they wanted their IRR higher, and in another because they felt if a specific metric was inflated considerably more, it would be easier to execute something. I'm not kidding around either, that happened.

Inflating some metric is one thing. Saying that a thing that's impossible by laws of physics, has been done in your product is quite another. I do admit however that it's not the first time. When I worked in a large tech company, our investment arm asked me to look at a startup that was clearly the same kind of fraud (doing things that are impossible according to Shannon), and indeed did turn out to be a fraud. But that's one example in my entire career.

Whether or not it’s “abstraction” really depends on who hears it. I don’t think it’s cool but maybe in a slightly more fuzzy area as to if it’s a lie.

Surely there would be a contract with the military if that was the claim and surely these investors did due diligence and audited the contracts, nobody would ever get wrapped up in the hysteria and hype and burn a billion dollars…

From what I saw from the trial, no one did any real diligence at all, except the people who testified they didn't invest because they did diligence. The Devos family office testimony was.... laughable, what a joke, they basically just went off what everyone else said, and that has primarily been my experience in startup. If big names are involved, people almost ALWAYS do the diligence they want to do so that you pass, not so that you fail.

tbqh, I came away from the trial really feeling like the investors who testified against her deserved what they got because of how they approached the situation. IMO Elizabeth Holms isn't some mastermind manipulator, she's more akin to a child who was enabled by moronic parents.

You don’t get to commit fraud just because your victims should have known better. That’s not how the law works

It seems so, as she's going to jail. However, that wasn't the crux of my original point. The legal standard and the SV standard have been divergent in may circles for quite some time. May folks do "take liberties with the truth" because your investors should know better - If someone went for coffee with 26th US secretary of defense for the united states of america and explained how it would be useful for the military, and that person said it would be and they could easily have the military look at testing it, I don't believe it would be uncommon for a founder to then turn around and say to an investor "the military is going to be running a test with our equipment", I wouldn't traditionally expect then to have the possibility of being prosecuted for wire fraud hanging over me if my startup failed. That said, we should all now _expect_ the legal standard is the SV standard. It will be interesting to see how it impacts the industry, if at all.

> If someone went for coffee with 26th US secretary of defense for the united states of america and explained how it would be useful for the military, and that person said it would be and they could easily have the military look at testing it, I don't believe it would be uncommon for a founder to then turn around and say to an investor "the military is going to be running a test with our equipment", I wouldn't traditionally expect then to have the possibility of being prosecuted for wire fraud hanging over me if my startup failed.

I think the crucial distinction between a slight stretch like you describe and what Holmes did is the tense: the military is going to test our product vs. the military is testing our product (which did not, uh, exactly exist then). One is a rosy forecast and the other is just a lie.

> ...say to an investor "the military is going to be running a test with our equipment"...

But as I understood it, that's not what she said. Your phrasing is about the future, i.e. something nobody can know at the time it is being said whether it will come true or not. Therefore, something anyone with the least bit of common sense would think "Yeah, let's hope that will happen" about.

Completely different than saying "the military is running / has run a test with our equipment". That's a statement of actual ongoing or concluded fact. If it actually is happening or has happened when you say it is happening or has happened, then you're telling the truth; if it isn't / hasn't, then it's a lie.

From what I gather, what she said was the latter -- but it never actually happened: So, a lie. Said with intent to induce people to give her money: So, fraud.

Fortunately the law doesn't have a victim blaming statute in these cases, so any mistakes the investors made don't excuse Holmes for lying and commiting fraud.

Yeah, this is serious stuff. It's not as if it were some piddling shit like rape, where victim blaming is all but mandatory.

You are completely uninformed on the details of the case. She absolutely did tell these lies to investors.

>She said it was being trialed by the military, in reality what had happened was Mattis said he would be able to introduce the technology to the military. This type of "abstraction"...

That's pretty far from a fair summary. Mattis made the introduction, Holmes could not get past the initial military diligence, and killed the project, and yet years later continued to represent that it was an important part of Theranos' business, to add an imprimatur of success and reliability of the testing.

> This type of "abstraction" is in very many pitches I've read over the years.

Nice of you to use quote marks... Because that's not at all what "abstraction" means.

> What Elizabeth did wasn't far from what many many many founders do.

Hard disagree on that one.

It's certainly not common nor acceptable to knowingly lie to investors like Holmes did.

There is nothing normal or common about the type of fraud that Holmes perpetuated, despite attempts to downplay or normalize it.

> I'm sure it's going to be an unpopular opinion but personally if this is the new standard, the SEC should have a pretty long list of founders to try at this standard.

Can you name a single one? Or is this just cynical speculation?

I've seen a lot of attempts to downplay Theranos' blatant fraud as an "everybody does it", but I haven't seen anyone calling out other companies who were caught committing fraud and everyone, including their investors, just shrugged it off. It's not normal.

Of course I can name multiple, I've been involved in hundreds and hundreds and hundreds of millions of dollars in startups deals, I've spent over 15 years in the industry...

Are you willing or can you share the names of some of these companies?

No. I don't think what they did was wrong a), and, b) because it genuinely is prevalent. Maybe if the statue of limitations passed on some of the startups I've been involved in, I might consider it, but even then, doubtful.. It doesn't matter what was done, the startups were successful and the investors got paid back, so who is going to ask questions and why?

Were any of these startups providing medical care or diagnostics to people on false pretenses?

She was found guilty of investor fraud, not medical/patient fraud.

Your implied question is "why is this case special?", when so many startups supposedly do this. Most startups aren't giving people input into major medical decisions.

I mean, on one hand, yes, if nobody is harmed, there really isn't a case for fraud.

But I agree with the other responder, I've also been in many startups and have certainly seen tons of exaggerations about the future but never outright lies about the present that Holmes was guilty of. Even worse, if you're going to lie about the present, you should at least have strong confidence your lie is physically possible - Theranos tech was always a sham from day 1.

I just don't buy at all the "everyone in Silicon Valley does it" spiel, because if they did, investors of failed startups (i.e. the majority) would have tons of reasons to sue and put people in jail.

Frankly, making an outrageous claim and then just blithely going "Nope, I'm not going to give even a single example" makes it sound like you are spouting just as much bullshit as Holmes did. Or alternatively, like you're just not understanding the difference between over-optimistic forecasts and claims of actual outcome. Which is it?

No. It's one thing to say "My company is going to change the world" or "I am the best president of all time". It's different to repeatedly make a claim that is verifiably untrue, like "I have perfected cold fusion" or "I have a cure for all forms of cancer". The average startup founder doesn't do this, no sane person does this.

> "The saddest thing of it all to me is that she was onto something, people are having success with her ideas today."

This is a very poorly informed statement. Who bought the valuable IP that Theranos spent hundreds of millions of dollars developing? No one is "having success with her ideas" because the ideas didn't exist. That's why she's going to prison

Exactly. Many of her ideas are physically impossible because there is good evidence capillary blood isn't homogeneous enough to give accurate results for quantitative tests at the volumes Holmes was touting.

Underrated comment. The investors were investing in her because she was willing to take the risk of move fast and break things (where things in this case is the diagnostic accuracy of blood tests). She got nailed, ironically, not for hurting patients, but instead for harming the egos of the investors.

As a participant in the startup investment world you are willing Participant of SUSPENSION OF DISBELIEF. This is roughly equivalent speak for REALITY DISTORTION FIELD. The point is you as an investor are a willing participant in believing that something more is possible than what has already been achieved.

If you're a founder: you should be worried about being held to a legal standard of the representations and warranties. That section is in every share purchase agreement and it's the stuff that gets you sued. And sometimes, like in this case, it's WIRE FRAUD and it's criminal. Crazy. We need another section to replace REPS AND WARRANTIES.

> We need another section to replace REPS AND WARRANTIES

Ain't that the truth. Hand on heart: the number of decks I saw last year that have an INSANE 2 page legal disclaimer at the start AND finish of the deck is up ~90%, ninety percent. I knew people who used this as a signal of an inexperienced founder, over night it's the signal of an experienced founder.

"Scaling too fast" has definitely a better vibe to it than "found guilty" :)

Are there any public example of the claims you are making?

Reply on news.ycombinator.com