YouSaid · the spoken record

Balaji Srinivasan

lines on the record
605
first
2022-10-20
most recent
2022-10-20
sittings or episodes
1
sources
podcast

Every line below is reproduced as it was said and linked to the record it came from. Nothing here is summarised or generated. Directory · Search · Corrections

  1. That sets these prices. And it is in theory only for Medicare, but all the private insurers key off of that. And AMA basically publishes a list of these so-called the CPT codes, which is like the coding, the biomedical coding of this. What each medical process is worth and whatnot. So it's like I don't remember all the numbers, but it's like a five digit code, and it's like, okay, I got a test for cystic fibrosis or a test for this or a test for that. God help you if your medical billing is erroneous. Why? The insurance company will reject it because it doesn't pay for that. This is this giant process of trying to encode every possible ailment and condition into these CPT codes and that you can literally get degrees in medical billing just for this enormous inefficient industry. Okay, like literally medical billing is a whole field.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  2. American Medical Association, CPT Current Procedural Terminology. Okay, basically, you know, Marx's labor theory of value where people are supposed to be paid on their effort. Of course, the issue with this is that you'd be paying a physicist to try to dunk and they'd be trying, but they wouldn't actually probably be able to do it. They'd be trying real hard. Whereas you actually want to pay people invasive results. Cheaply attained results are actually better than expensively attained non-results, perhaps obvious. Nevertheless, the way that the U.S. medical system has payouts, it's based on so-called RVUs, relative value units. And this is something where there's a government body

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  3. And that is the most nimble thing ever relative to FDA's design review. So just to review, A, talked about how FDA was blocking all this stuff. B, talked about why ambulatory medicine is capitalist medicine, ambulance medicine is socialist medicine. C talked about how with the diagnostic stuff, we can shift it over to ambulatory. D talked about how there's lots of things where you could have a combination of doctor and AI in an app that you kind of quickly self-diagnose. Some of this is happening now. some of the telemedicine laws were relaxed during COVID where now people, you know, a doctor from Wyoming can prescribe for somebody in Minnesota. Like some of that stuff was relaxed during COVID. There's other broken things in Melchism. I'll just name two more and then kind of move on. I mentioned AMA and CPT.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  4. There's a combination of both American doctors and the FDA that team up to prevent this or slow this. And one argument is, oh, the AI is not better than a human 100% of the time because it's not deployed yet. Therefore, could make an error, therefore, it's bad. Even if it's better than 99% of doctors, 90% of the time. Another argument is the software has to go through design control. Now, basically, once you understand how FDA works, basically imagine the most bureaucratic, frozen process for code deployment at any company ever.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  5. Okay, so you want this amazing, amazing doctor there, right? How could we get that? I mentioned the app that doesn't exist, which is like a better version of the 3500B, right? Here's another app that doesn't exist, and this is one that FDA is actively quashed. Why can't you just take an image of a mole or something like that with the incredible cameras we have, a huge amount of medical imaging should be able to be done at home? And it goes to doctors, whether it's in the US or the Philippines or India. I mean, teller radiology exists. Why can you not do that for dermatology, for everything else? You should be able to literally just hold the thing up and with a combination of both AI and MDs, just diagnose. That should exist. Answer is

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  6. And just like this Larry Smar guy that I mentioned, the measured man, you might be able to shift more and more things to ambulatory. And one of the things about this also is the medical system is set up in a bad way where the primary care physician is the one who is not the top of their class, but the guys who are at the bottom of the pinball machine, the surgeons and the radiologists, once your stuff is already broken. They're the ones who are paid the most. So a lot of the skill collects at the post break stage. Where you actually want Googie Hauser MD is at the upstream stage.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  7. Savings count, right? The thing about this is once you realize, first, ambulatory medicine is capitalistic medicine, ambulance medicine is socialist medicine. You want to shift people more towards ambulatory. Guess what? That's in their interest as well. Now that brings us back to the monitoring, right? The continuous monitoring, whether eventually it's Mike Snyder's integrum, the V1 is the quantified self and the Apple Watch and the continuous glucose meters. And a VN is the Mike Snyder integrum. There's a site called Q.Bio, which is doing this also, right? Eventually this stuff will hopefully just be an advice that just measures tons and tons of variables on you, right? There's ways of measuring some of these metabolites without breaking the skin. So it's like it's not, you don't have to keep breaking the skin over and over. There's various ways of doing that. So now you've got something where you've got the monitoring, you've got the dashboards, you've got the alerts.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  8. Whereas if you're coming in with an ambulance, then people don't want to be turned down, and I understand why. What this suggests, by the way, is that you should only have insurance for the edge cases. Insurance should only cover the ambulance, not the ambulatory. Most people are losing money on insurance, right? Because most people are paying more in in premiums than they are getting out. It's just that this huge flight of dollars through the air that no one can make heads or tails out of, oh, the other aspect that's obviously broken is employer provided health insurance, which just started after World War II. So, you know, auto insurance is in a much more competitive market. You don't whip out your auto insurance card at the gas station to pay for your gas, right? You only whip it out when there's a crash, right? That's what, quote, health insurance should be. And the Singapore model is actually a pretty good one for this, where they have sort of a mandatory HSA. You have to like put some money in that and that pays for your healthcare bills, but then it's cash out.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  9. Okay, with a dermatologist basically turning you down because you don't have enough money and you go to another dermatologist because you can comparison shot there. It's not usually an emergency, right?

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  10. Basically, the most capitalistic areas of medicine are the most functional areas of medicine. So that's, say, the places where you can walk in and walk out, whether that's dentistry, dermatology, plastic surgery, even veterinary medicine, which is not human. Where you can make a conscious decision say, okay, I want this care or I don't want this. I see a price list. I can pay cash, right? If I don't like it, I go to another dermatologist. There's few dermatological emergencies. That's why dermatologists have a great quality of life. By contrast, when you're being willed in on a gurney, you need it right now. And you're unconscious or what have you or you're not in a capacity to deal with it, right? And so these are the two extremes. It's like ambulatory medicine. You can walk around and pick. And like ambulance medicine. Okay. And what that means is the more ambulatory the medicine, the more legitimate capitalism is in that situation.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  11. themselves on a gurney when they're being wheeled in and now somebody at their moment of vulnerability is charging this insane price for their care and many people in the US have had this horrible experience where they're bankrupted or scared of being bankrupt of medical bills. Therefore, the concept of adding more capitalism to medicine scares them and they think it's horrible and you're some like awful greedy tech bro kind of thing, all right? Let me say I understand that concern and let me you know kind of let me pull tease that apart a little bit, right?

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  12. So basically, part of it is decentralizing control back to the individual, right? Now, I've talked about FDA at length, but let me talk about some other broken parts of the US system, right? There's like AMA, there's CPT, there's CPOM, there's this all these regulations which see normally in capitalism you have a buyer and a seller, right? Duh. In medicine, you have third-party regulation and fourth-party pricing and fifth-party payment. Okay, so third-party regulation, FDA is regulating it. Fourth party pricing, it is the CPT codes, right? Fifth party payment, it's the insurance companies, right? And just to discuss these bits of the system, first, why are some people against capitalism and medicine? I actually understand why they're against it, because they are visualizing the

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  13. They don't like paying to get back to normal, but they will absolutely pay tons of money to get better than normal. They'll pay for fitness, they'll pay for makeup, they'll pay for hair, they'll pay for this and that, right? So that's actually the back door, and you can do tons of things there where obviously being healthier is also protective. You can actually show the studies on this. So this way you build out all the tooling to get healthier, and that actually helps on this axis. Few of the things of US medical system while I'm onics, you got me on this topic.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  14. But why is this the backdrop of medicine? Because since FDA only regulates those things that are meant to diagnose and treat a disease, All the stuff that is meant to improve an otherwise healthy person is potentially out of their purview. Supplements are one interesting aspect that they were carved out in the mid-90s. And that's why the supplement industry is big, because FDA doesn't have as tight a rein on that. But all of the Fitbit CGM, continuous glucose meter type stuff, you can crank out all kinds of things that help people get fitter that will also actually have just general health value, but you're not quite marketing them to diagnose or treat a disease. See what I'm saying? You're marketing them for the purpose of fitness. This is a market. Why? Because psychologically people

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  15. So, what that means then is that intervention of that 15 or 30 minute appointment with the doctor, whatever drug they prescribe, better hit you like Thor's hammer to put you back on the straight and narrow because that's only with you for like a few seconds, you know, a few minutes or whatever. The doctor's only with you for a few minutes. The drug is only, you know, some drugs are very powerful, so they actually do work like this, okay? But your fitness is your own responsibility, and that's a continuing forcing function every day. And again, we get back to decentralization, right? The decentralization of responsibility from somebody thinking themselves merely as a patient to an active participant in their own health who's doing their own monitoring of their own health, right? And logging all their stuff, who's eating properly and looking at the effect of their diet on things like their continuous glucose monitor as a V1, but other things, right? Who is as fit as they can possibly be? Like these are kind of obvious things.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  16. And so your fitness, your diet, that's your responsibility. But when you go into a doctor's office, suddenly it becomes lie back and think of England. Okay, suddenly you become passive. Suddenly, oh, your doctor is Dr. Google. Well, your doctor must be a moron. You're going and trying to take care of yourself. You're Googling symptoms. Oh, how stupid you are. I have a medical degree. And that doctor, see, the thing is, if you come in and you've self-diagnosed or you've done some of your own research, if you're right, and if they've got an ego about it, they're undermined. And if you're wrong, they're like, you know, ha ha, you know, arrogant. But either way, if they've got this kind of mindset, they have an incentive to resist the patient taking care of themselves. Isn't that the doctor's job, right? And they're kind of taught to behave like this, many of them.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  17. Yeah, the ultimate medicine, right? So if they could just prescribe the effects of it, it's just like, boom, just massive effect, right? Like you're fitting enough, you do the resistance training, it helps with preventive diabetes. Every kind of thing in the world, you see a significant treatment effect. Yet your fitness is your own responsibility. You go to some gym, 24-hour fitness. What do they have? They have on the wall exhortations like your body is your responsibility, right? Am I right? And Go ahead, but

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  18. A lot of the talent has left the building, but was there. C, we're entering the decentralizing era. And D, you know, like be like Satoshi. Satoshi founded Bitcoin because he knew you could not reform the Fed. There's everybody's trying to go and reform, reform, reform. The reason trying to reform is we haven't figured out the mechanism to build something new. And now perhaps we have that. So I've named a few of them, right? I'll name one more. Relate to utterance. Fitness is actually the backdoor to a lot of medicine. Why is that? You go to any conference. It could be neurology. It could be cardiology. You'll find somebody who's giving a talk that says something along the lines of fitness is the ultimate drug.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  19. If, and a lot of people are drawn in like moths to the flame for these titles of the establishment, I want to be head of this, I want to be head of that, right? And really what it is, it's like. Know becoming head of Kazakhstan in the mid 1980s in the Soviet Union, the Kazakhstan SSR, right? Soviet Socialist Republic, before the thing was going to like crumble potentially, right? In many ways, it's becoming folks who are just totally status obsessed getting these positions. But like a lot of the merit, all the folks with merit are kind of leaving the government and going into tech or crypto or what have you, right? Even if these agencies were hollow before in some ways, they're becoming hollower because they have less talent there, right? So A, you can't hire and fire very easily. You can hire a little bit. You can't really fire.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  20. And if you endanger that refueling, that annual budget renewal or what have you, then the whole agency will basically be like, okay, you're bad and you can be disciplined or sometimes with rare except you can be booted. But what that means is that FDA commissioner is actually a white elephant. It's a ceremonial role, really, right? The term white elephant is like basically the Maharaja gives you a white elephant as a gift seems great. Next day it's eaten all of your grass. It's pooped on your lawn. It has like just put a foot on your car and smash it. But you can't give it away. It's a white elephant. The Maharaja gave it to you, right? That's what being like FDA commissioner is. It's the kind of thing where.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  21. So, the one way you can truly screw up within a regulatory bureaucracy is if you sort of endanger the annual budget renewal. Think of it as this mini Death Star that's coming to dock against the max death star for its annual refuel. And it's talking about all the corporate criminals that it's prosecuted, the quotas, like the police quotas, the ticketing, you know, and If they don't have a crisis, they will invent one. Just again, just like TSA, just like other agencies, you're more familiar with you can kind of map it back. Look at the guns and drugs we've seized. And say you have an incentive for creating these crises or manufacturing them or exaggerating them.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  22. This whole process where they get to appeal it and so on and so forth. And they're sitting in the office while you're trying to fire them. And they're complaining to everybody around them that this guy's trying to fire me. He's such a bad guy, blah, blah, right? And everybody around, even if they may think that guy is doing a bad job, they're like, wait a second, he's trying to fire you. He might try to fire me too. And so anybody who tries to fire somebody at FDA just gets a faceful lead for their troubles. What they instead will do is sometimes they'll just transfer somebody to the basement or something so they don't have to deal with them if they're truly bad. Okay. But the thing about this is there is only one caveat Douglas factor number eight. The notoriety of the offense or its impact upon the reputation of the agency. There's that word again, reputation of reputation and power.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  23. So So basically, see, a CEO of a company, well, it's very difficult. They can hire and fire, right? So, in theory, they can do surgery on the organism. And like, you know, Steve Jobs took over Apple and was able to hire and fire, raise money, do this, that he basically had root over Apple that he was a system administrator, right? He had full permissions. As FDA commissioner, you do not have full permissions over FDA, let alone like the whole structure around it, right? If your FDA commissioner, you are not the CEO of the agency. Okay, lots of these folks there have career tenure. They can't be fired. They can't even really be disciplined. There's something called the Douglas Factors. You ever heard the Douglas factors? It's like the Miranda rights. for federal employees. Okay, you have the rights for man says it. So basically, if you've heard that federal employees can't be fired, the Douglas factors are how that's actually operationalized. When you try to fire somebody,

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  24. You could go back and do that study across lots and lots of different drugs, and you probably find quite a lot. Alex Tabarak and others have written on this, right? Daniel Heninger has written on this. That's just like one example. I mean, I gave the pandemic example. The fact that they held up the EUAs for the tests and didn't do challenge trials. That's like, you know, that's like a million American dead that could have been orders of magnitude less if we had gotten the vaccine out to vulnerable populations sooner. So you're talking about something that has a total monopoly on global health. You can't know what it is without that unless you have zones that are FDA free but that have some form of regulation. As I mentioned, it's a V3. It's not going back to zero regulation everybody's self, but it's a more reputable regulator just like Uber is a better regulator than the taxi medallions, right?

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  25. FDA is killed way more people than Theranos has. All right. Way more. Just take drug lag alone. Whenever you have a drug that works and reduce morbidity and mortality after it is actually generally available, but was delayed for months or years, the integral under that curve is the excess morbidian mortality attributable to FDA's drug lag.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  26. Right So, I mean, one of the things about this, by the way, is because there are so many movies made about Darnos, that's one of the reasons why people have sort of been scared off from doing diagnostics to some extent. Why? Because visas are like, oh, is this another Theranos? The diligence and everything, everyone's looking at it. Oh, blood testing, one drop of blood, huh? Hurts with recruiting. Essentially, a lot of the media and stuff around that basically has pathologized the thing that we want to have a lot more entrance in, right? Now, you know, one way of thinking about it is.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  27. Takes blood, runs every test, right? Maybe that's a few thousand dollars a year. Maybe eventually gets to a few hundred dollars a year. And that's expensive in some ways. But boy, that's better health insurance in other ways.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  28. Why? Because we need a lot of sunlight. So often inside you're tapping on your screen. So, what do we do? Take actually significant vitamin D infusions. That's like a small example of where baselines differ between people. So continuous diagnostics, what could that mean? That could mean things like the continuous glucose meter, it's quantified self, it's like continuous blood testing, right? So you have a so-called mobile phlebotomist. This is something which phlebotomists takes blood, right? Mobile phlebotomy would come to your office, come to your remote office. This is a great business for people. I think, you know, you can revisit this in 2022. People tried this in the 2010s, but I think it's worth revisiting. Mole phlebotomist comes every week or every month

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  29. And moreover, he could take, you know, I don't think he did this, but if you took treatments, if you took drugs, right, you could actually show what your steady state was if you tracked over time, show what your disease state or six state was, and then this drug pushes you back into non-disease state. You can actually get a quantitative readout of what, you know, like steady state was, right? And that steady state, you know, your expression levels across all these genes, your small molecules, basically everything you can measure, that's going to vary from person to person, right? What's healthy and natural for you may be a different baseline than for me. For example, people who are a small example, people who are South Asian or have dark skin tend to have vitamin D deficiency.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  30. Like symptoms, if I'm remembering it accurately. So you can see, oh, wait a second, these are things that are that I can see in my readouts that I would only have the vaguest interpretation of as like a human being, right?

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  31. Actually, kind of insane to think about this stuff that is all the other side of the world 10,000 miles away, but a few millimeters away, you don't really know what's going on, right? And that's starting to change with all the quantified self devices, the hundreds of millions of Apple watches and Fitbits and stuff, right? You're also starting to see continuous glucose meters, which are very important. They're starting to give you readouts. People are seeing, wow, this is spiking my insulin, or rather this is spiking my blood sugar. And it might be something you didn't predict. It varies for different people. For some people, banana isn't a big deal for others. It's actually quite bad for the butcher. What happens when you extend that? About 10 years ago, a guy Mike Snyder, professor at Stanford, did something called the Integrome, where he just threw the kitchen sink of all the diagnostics he could at himself over the period of, I think, a few weeks or a few months or forget the exact duration. And he's able to do things where he could see during that period he got a cold or something. And he could see in the expression data, the gene expression data, that he was getting sick before he felt sick. He could also see that something about that, like viral infection made him develop diabetes.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  32. Then it turned out to be an early warning of a serious condition, and you had to, I think, go for surgery or something. And he was starting to think, well, look, the way that we're doing medicine right now is it's not quite pre-germ theory of disease, but it is pre-continuous diagnostics. Continuous diagnostics, just to talk about this for a second, I mentioned one angle on which you go after FDA, which is like the better phase four, right? I've mentioned the concept of better reviews in general. I mentioned Vita DAO, which is like a community that is going after longevity. Let's talk about continuous diagnostics. So basically, we know better what is going on in Bangalore or Budapest than in our own body.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  33. So, you can see whether things are breaking, and then you deal with it just like you've got server uptime things. You don't wait necessarily for the site to go down. You start seeing, oh, response rates are spiking. We need to add more servers, right? You have some warning, okay? Even 10 years ago, there's this article called The Measured Man in the Atlantic where this guy physicist Larry Smarr, okay. What he was doing is he was essentially doing a bunch of measurements on himself. And he was finding that there were predictors of inflammation that were spiking. And he went to the doctor, showed the charts, and the doctor was like, I can't do anything with this.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  34. That will say, look, we want this new drug or we want this new treatment or we want this new device and we're willing to crowdfund 10,000 units. So please give us the thing and we'll write a very fair review of it and we'll also all evaluate it as a community and so on. So you turn these people from just passive patients into active participants in their health. That's a community part. And they've got the kind of biomedical technical leadership there. Now what is the kind of prototype of something like this? Something like VitaDAO is very interesting. Things like molecule DAO are very interesting. It'll start with things like longevity, right? And why is that? Because the entire model of FDA, this 20th century model, is a way for somebody to have a disease and then try to cure them. Versus, you know, saying an ounce of prevention is worth a pound of cure, right? Why are we not actually tracking folks and getting a constant dashboard on yourself?

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  35. Okay, just like there are props who defected from computer science academia to become Larry and Sergei and whatever, or they weren't props, they were grad students, right? In the same way you'll have defectors who have the credentials from the old world, but can build up the new. Just like there's folks from Wall Street who have come into cryptocurrency and helped legitimate it, right? Just like there's folks who left Salzburger to come to Substack. We have these folks who, by defecting, they help, and then they're also supplemented by all this new talent coming in, right? That combination of things is how you build a new system. It's not completely by itself, nor is it trying to reform the old at some fusion. So in this new system, who do you have? You have like the most entrepreneurial and innovative MDs. You have the most entrepreneurial and innovative professors. And you have the founders of actual new products and stuff. And they are giving open source reviews of these products. And they're also building a community.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  36. Just start with that. That's pure information. You're under free speech. That's still the most defended thing. Literally, just publishing reviews. And not just reviews by any old person. It turns out that FDA typically will use expert panels, where there's expert panels. It's like professors from Harvard or things like that. So what that is, is this concept of reputational bridge? What you want to do is you want to have folks who are, let's say, biotech entrepreneurs or they're profs like Sinclair or what have you. You do want to have the reviews of the crowd. But you also want to have, especially in medicine, by medicine, you want to have the reviews of experts of some kind. There's going to be defectors from the current establishment.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  37. I mean, it's basically the same thing. Airbnb could use competition. I think that it would be good to have competition for them and there are other kinds of sites opening up. But the fundamental concept of the cloud regulator now, let's apply.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  38. Because that exit is now possible. Now, you may argue, oh, well, lift a new bird, they're not profitable anymore. And there's two different criticisms of them. One is, oh, they're not profitable. Or, oh, they're charging too much. And I think part of this is because of certain kinds of the regulatory status caught up to try to make them uncompetitive. For example, they don't allow people in some states to identify themselves as independent contractors, even if they are part-time. There's various other kinds of rules and regulations. You know, in Austin for a while, Uber was even banned, what have you, right? Netnet, though, like Uber, grab, Gojack, Lyft, DD, like ride sharing as a concept is now out there. And whatever the next version is, whether it's self-driving, like while it's like a very hard fought battle in the regulatory state keeps trying to push things back into the garage, this is a fundamentally better way of just doing regulation of taxis. Similarly, Airbnb for hotels.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  39. Uber is the new regulator, and Uber drivers are the drivers. Lyft is the competing regulator and Lyft drivers are the new drivers. So you have a regulator driver fusion versus another regulator driver fusion. You no longer have a monopoly. You have multiple parties. Okay, you have a competitive market. This is the concept of like polycentric law, right? Where you have multiple different legal regimes in the same jurisdiction overlapping that you can choose between with a tap of a button, right? All these concepts from like libertarian theory like polycentric law or all these things are becoming more possible now that the internet has increased microeconomic leverage.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  40. Okay, because they were the persistent actors in the ecosystem. Taxi riders had nothing in common, didn't even know each other. In New York, some guy gets in a taxi, another guy. They had no way to communicate with each other. So the persistent actors in some of the regulators and the drivers, and they had this cozy kind of thing and medallion prices just kept going up. And this was the sort of collaboration on artificial scarcity. Afterwards with Uber and Lyft and other entrants, you had something interesting, a different kind of regulator driver fusion. If you assume regulatory capture exists and lean into it,

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  41. Does that do? Think about how much better that is in taxi medallions rather than a six month or annual inspection, you have reports from every single writer. Before Uber, it was the taxi drivers and taxi regulators were in a little monopoly locally.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  42. From Mexico to help somebody in Moldova or vice versa because it's fundamentally international those ratings that you have a network effect. And there's another aspect to it, which is these are better regulators than the land regulators. For example, Uber is a better regulator than the taxi medallions. Why? Every right is GPS track. There's ratings on both the driver and the passenger side. Both parties know that payment can be rendered in a standard currency. If you have below a certain star rating on either side, you get deplatformed and so on. to protect either rider or driver and on and on right

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  43. So, the thing is, the model we've talked about is the land regulator of a nation state and a territorially bounded thing. But the cloud regulator, what's the cloud regulator? That is Amazon Star ratings. That's Yelp, that is eBay, that is Airbnb, that is Uber and Lyft, and so on and so forth. It's also actually Gmail and Google. Why? Because you're doing spam filtering and you are doing ranking of mails with priority inbox, right? With Google itself, they ban malware links. So the bad actors are out and they're ranking them. How about Apple, the App Store, right? They ban bad actors and they do star ratings. When you start actually applying this lens, PayPal, they've got a reputation. Every single web service that's at the scale of like tens of millions or hundreds of millions of people has had to build a cloud regulator. And the crucial thing is it scales across borders. So you can use the data.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  44. You don't want to go to Starbucks and put a dipstick into every coffee to see if it's poisoned or something like that. You sort of want to enter a zone where you know things are basically good and you pay that one diligence cost on the zone itself, right? Whether it's a digital or physical zone. And then the regulator's taking care of it and they've baked in the regulatory cost into some subscription fee of some kind, right?

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  45. When I talk about exit the Fed, that's the crypto economy, right? What does exit the FDA look like? Well, one key insight is that many of the big scale tech companies can be thought of as cloud regulators rather than land regulators. What do I mean by that? Well, first, what is regulation? People do want a regulated marketplace. They want A, quality ratings, like on a one to five star scale, and B, bands of bad actors like the zero star frauds and scammers and so on. And these are distinct, right? Somebody who's like a low quality but well-intentioned person is different than a smart and evil person. Those are two different kinds of failure modes you could have in a marketplace, right? Why is it rational for people to want a regulated marketplace, especially for health? Because they want to pay essentially one entry cost and then they don't have to evaluate everything separately where they may not have the technical information to do that.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  46. So they have a bad yelp for drugs. Could we make a better one? We could make a better one. Just modern UX. The key insight here, by the way, which is a non-obvious point. And I've got a whole talk on this, actually, that I should probably release. I actually did like almost eight, nine years ago. It's called Regulation Is Information. Product quality is a digital signal. Okay, what do I mean by that? Basically,

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  47. Train crash of all the things that you're paying for that you supposedly think are good. As I mentioned, one response is to go QAnon and people say, oh, don't trust anything. But the better response is decentralizing FDA. Okay. So I will say one other thing, which is I mentioned this concept of improving the Form 3500B, where you scan, go ahead.

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  48. What has changed? What's changed is in the 2010s, late 2000s and 2010s and so on, with the advent of social media, with the advent of a bunch of millionaires who were independent, with the advent of Uber and Airbnb, with the advent of cryptocurrency, with the diminution of trust in institutions. It used to be really taboo to even talk about the FDA as potentially bad in like, you know, 2010, 2009. But now people have just seen faceplant after faceplant by the institutions and people are much more open to the concept that they may actually not have it all together. And I think it's, you know, you probably see some tracking poll or something like that, but I wouldn't be surprised if it's like a 20 or 30 point drop after the CDC failed to control disease and the FDA failed and the entire biomedical regulatory establishment and scientific establishment was saying masks don't work before they do. This was just a

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  49. A culture of acquiescence towards the FDA. The FDA was their primary customer. So just like in a sense, it's rational, you know, Amazon talks about being customer obsessed, right? What Rob did was rational for that time, right? What G Kirp Rob did was saying our customers, the FDA, that's our primary customer. Nobody else matters. They are satisfied first. Every single trade-off that has to be made is FDA, right? And really, that's why the two most important departments at many pharmaceutical companies arguably all are regulatory affairs. So, this entire thing is just like college education, these things may at some point have been a good concept, but the price has just risen and risen and risen until it's at the limit price and beyond. So

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source

  50. Where it's a piece of paper so you don't see the gun attached to it. But if you were to defy that, it's like Grand Theft Auto, where you get one star, two star, three stars, four stars, five stars, and eventually, you know, you have some serious stuff on your hands. So once you understand that every law is backed by force, like that Brazilian guy with the machine gun that Rob mentioned, these guys are the regulatory police. See, for a time, what happened was... You had the captured industry because all of the folks who were in pharmaceuticals were, as Carpenter said,

    2022-10-20 · Lex Fridman Podcast · #331 – Balaji Srinivasan: How to Fix Government, Twitter, Science, and the FDA · IDENTIFIED FROM THE TRANSCRIPT · source