Contrarian question: is Elizabeth Holmes really as bad as consensus makes her out to be? I read Bad Blood when it came out. Here are some questions I had: 1.

Contrarian question: is Elizabeth Holmes really as bad as consensus makes her out to be? I read Bad Blood when it came out. Here are some questions I had: 1. Base rate insensitivity. Carreyrou argues that Theranos prototype(s) were abundant in Type 1 and Type 2 errors. Yet, as far as I recall, it was never specified what the acceptable error rate is for a blood testing machine, and precisely how deviant Theranos' machines were. If we found out, for example, that an acceptable Type 1 error rate for a blood machine was 1%, and that Theranos' machines were yielding 3% Type 1 errors, I wouldn't be as offended had they been yielding 20% error rates. So exactly how error prone were Theranos machines compared to base rate? It was never specified. This makes it really hard to quantify the level of harm they were (or might have) imposed on society. 2. Theranos might have had a legitimate path to product. No one seems to dispute that Theranos had actual prototypes that functioned to some degree. The issue is that they functioned (i) with unsuitably high error rates (see above) and that (ii) Theranos' machines were on their own only able to test for certain things. But it's still unclear to me if they had an actual path to finishing their machines to spec. (The book concentrated on Holmes' misleading people on how much progress they had made, but this is distinct from whether or not Theranos had a secret plan to finish things). For sake of argument suppose Theranos needed another 3 years to finish their machines to spec. If Holmes had pulled it off, we might had called her a hero (at least I might have). So did Holmes have a path? It's unclear reading the book, from what I recall. Of course, the history of SV is full of people pushing the edges. This leads us to point 3: 3. Carreyrou's anecdotes of harm were weak. Carreyrou was unable to find compelling cases of people being severely harmed by Theranos machines, or so one could argue. Here "severe harm" to me means: failure to identify life threatening diseases (Type 2) causing, i.e., death. If I recall Carreyrou did find one case of Theranos machines misdiagnosing severe diseases (Type 1 error causing someone thousands of dollars of further blood tests + serious anxiety along the way). Of course, that's not a good thing. I wouldn't wish that upon anyone. But only one case? (Keep in mind that normal blood tests already can give people Type 1 errors [causing anxiety, further medical expenses, etc]; see point 1). And compare this harm to the amount of good Theranos would had been able to add to society had they succeeded in their engineering efforts (assuming they did have a path to victory). 4. Holmes had skin in the game. If I recall, Holmes fought tooth and nail to maximize her equity in Theranos, and when the (alleged) house of cards came down, she had less than $1M in net worth. This strikes me as consistent behavior with someone who genuinely believed in her plan (and hence the long-term value of Theranos). At worst Holmes was totally delusional, but it seems less likely to me that she was a legitimate con artist. I just want to clarify: it's been a while since I've read Bad Blood (and actually, I quite enjoyed the book). And I don't condone misleading investors, fraudulent behavior, etc. I'm sure I'm missing some serious things Theranos did that could have been criminal. But Carreyrou (himself not an innovator) seemed at times to do a poor job considering the alternate viewpoint sincerely: that Holmes was just pushing the edges, and that it was at the end of the day worth it. His book no doubt set the dominoes in motion for Theranos to implode. So I felt compelled to explore this alternative viewpoint and to get it in writing. Cheers.