The Vergecast - Apple and Google are building a coronavirus tracking system into iOS and Android

Episode Date: April 10, 2020

Nilay Patel talks with The Verge's Adi Robertson, Casey Newton, and Nicole Wetsman about Apple and Google announcing a system for tracking the spread of the new coronavirus, allowing users to share data through Bluetooth Low Energy (BLE) transmissions and approved apps from health organizations. Links: Apple and Google are building a coronavirus tracking system into iOS and Android What is contact tracing? Why Bluetooth apps are bad at discovering new cases of COVID-19 Learn more about your ad choices. Visit podcastchoices.com/adchoices

Support for the show comes from Retool. Too many companies run critical operations on duct taped spreadsheets, Slack workflows, and whatever else they could cobble together. Not because they want to, but because building internal tools means weeks of waiting on someone else's backlog. That's where Retool comes in. Build custom internal tools just by describing what you need. Prompts something like,

Build me a revenue dashboard on our Salesforce data. And Retool actually builds it on your company's data, in your cloud with enterprise security built in. Go to retool.com slash vergecast. We all need to retool how we build software. Hello and welcome to an emergency episode of the Vergecast. I'm Neely. I'm joined by Addy Robertson.

Hi, Eddie. Hi. Nicole Wetzman. Hello. And Casey Newton. Hi, Neal. So I've assembled this illustrious group of Verge reporters because Apple and Google announced something pretty surprising today.

They are teaming up. You don't usually hear about Apple and Google. teaming up, but they are teaming up to build a set of APIs and technologies into iOS and Android that use Bluetooth LE to allow for something called contact tracing, which will eventually enable us to more accurately track the spread of the coronavirus. That's a lot. That's a lot of new vocabulary words. It's a lot of novel uses of technology. And it is a big deal because certainly if you've been listening to the Vergecast, and probably anyone, you're aware that Apple and Google are the two dominant mobile phone operating system vendors in this world.

So if they decide to do something together, the opportunity for it to be widespread and actually effective is higher than some of the other apps we've seen. But I want to start at the very basics, which is actually not about Bluetooth, not about their plan, but what contact tracing is. So Nicole, you're our health reporter. You've been covering this stuff. You actually just wrote the explainer. Walk people quickly through what contact tracing is and why it's important.

Yeah, so contact tracing is one of the major tools that public health officials have to manage any sort of infectious disease outbreak, whether that be something that reaches a pandemic level or is just, you know, a cluster of measles cases. And the way that it works is public health officials are trying to identify every single person who has whatever disease you're looking for. And then they're trying to figure out who that person came into contact with over the past, you know, set, period of time, who they might have been able to pass the disease onto who may be feeling sick, who might not be feeling sick yet. And then if any of those people get sick, they trace all their contact, so on and so forth. And so basically you're tracking the chain that a virus could travel from person to person and trying to break all of those possible change of transmission. So, for example, if I was diagnosed with COVID-19, a public health expert could come to me and

say, Nicole, who have you interacted with in the past week and a half? And I would tell them, you know, people that I live with or people that I saw at work or anyone that I interacted with at a party or something like that. And then the public health official would go to that part, all of those people and say, hey, you need to monitor yourself for symptoms of the illness or you need to quarantine for a period to see if you develop that illness. And that's how we stop the virus from being able to travel kind of around a community. So that is a very manual process. It's one that has been used before.

We kind of know how effective it is. Yes, it is like basically old, very old school detective work, kind of the bread and butter of a lot of public health response. That also seems like one of those things where you could just say, everybody has a computer in their pocket with location signals. We could just do this very easily. I mean, I've just heard that a lot. But there are reasons we haven't done that in the past and that we might do it now?

Well, we haven't really been forced to do contact tracing at an extremely, extremely high level, at least in the U.S. until this pandemic. You know, contact tracing has been a very key part of measles outbreaks, but those are typically smaller clusters. And we have the resources and personnel at public health agencies in order to kind of do that detective work in a very sort of, granular on the ground sort of way. We don't have the resources to do case tracking the way that we need to right now for a couple of reasons. One, we have a lot of cases because things got very out of control because our public health response was bad. And also systematic defending of public health agencies mean that we just don't have the people that we need to do stuff like this. So it sounds like we've got this huge volume. Computers are pretty good at database stuff, I would say.

That's a fair statement. I feel like you could argue with me about. it, but I feel strong in saying computers are good at database stuff. Addie, what is the kind of system that was announced today that Apple and Google want to use to enable contact tracing? So like you mentioned, Apple and Google are using a really, they're basically building a very specific version of a contract tracing system that uses Bluetooth low energy. So the idea is that there are two phases of this program. In the first phase, they build this API that public health authorities can use for apps

that they build. So you download an iOS or an Android app, and then it scans for other devices that are basically using the app that are within six feet, which is the distance that you are supposed to keep yourself to stay safe. So if you're within six feet of somebody and your phone is near their phone, then they exchange a sort of key handshake. And these keys are anonymous, but you just have a record of all of the keys you've had and generated. And that stays on your phone. That stays on your phone. And then if you develop symptoms and you get tested and you go to a facility and you test positive,

then that information is uploaded in a way that we're not totally sure about yet. And then the system will flag anyone who has interacted with your key, who has a sort of key that's linked to you. And then they'll say, hey, look, you were exposed to this person. They won't tell you who it is. They'll just say you were exposed. And you should do whatever this app is presumably telling you to do, which would probably be to self-quaranty. Wait, so, Adi, just to be clear, there's nothing about this that tracks your location.

Not remotely in the way that people usually mean that. It doesn't use GPS. It doesn't triangulate your location through, like, IP addresses, anything like that. So it cannot tell where you are physically. all it can do is check, okay, Bluetooth signal. Is there another Bluetooth signal from another phone? Yes? Okay, let's take a key from that phone.

You could argue that maybe there is a system in which you could sort of cross-reference these things on a massive scale to create correlations that would reveal your location. But in the way that, say, Google can tell when you visited a bodega, it cannot do that. So that's important. I think that's another reason they let Apple kind of go first. But I think the first question I got was, now everyone's going to know where I am. And it doesn't seem like this system on its face is designed to reveal that information. Way less than almost any, like, app you would install on your phone. Right.

So the split here that I think is just really important to underline is what Apple and Google have announced is, first, a set of APIs for apps to use. Other people have to go build those apps, public health officials, it sounds like, states, governments. They will build those apps. They will light up some sort of new Bluetooth beacon functionality. phone, then Apple and Google are going to maintain the sort of tracking of keys software on the phone and then the server that has all the, in the diagram, they're called diagnosis keys, which is extremely like a video game, in my opinion. But some server will have all the keys of people who have tested positive.

Every so often the server will tell all the phones, here are the keys that say positive tests, and your phone will light up if you have a key on your phone that matches that key. but they're not building any user interface as near as I can tell. And they are not building any sort of facility for tracking who tests positive. At least in the first phase. So the second phase of this is that they want to build this straight into iOS and Android. And they want as far as we can tell and everything I say from now on gets a little bit hazy. They want to add it as sort of a baseline option the way you can turn on like location tracking.

and you would opt into it in your menu or something, but you wouldn't have an app there by default. And in some way, you could opt in, and then it would do the handshake. And if you were exposed, it would somehow tell you, hey, look, you should download an app or something and be told that you were exposed.

I sound weird saying this because I'm not totally sure what exactly they're planning. But the basic idea is that it's something that's just going to be built into iOS and Android, so it's really easy for people to opt in. And if you actually want to or need to interface with it at any sort of higher or more complex level, then you can install the app. And the app would come from your government in some way, it sounds like. It seems like.

Like your local government. The reason I'm just like pushing on this distinction is it seems really important to note that Apple and Google are building the actual tracing functionality, but they're not building anything you need to use it. As far as we know. They're not building like Google Maps for coronavirus. Although Google is kind of doing that like over there. That's like another initiative.

But this idea that if you test positive, all of the phones you've been around should somehow get a push alert that those phones have been near somebody who tested positive, that's what people want. And this is the underlying mechanism of doing that. But it is not the user interface. It is not the testing protocol. It is not the validating of test results. So people don't troll the system with fake positives. there's a whole layer of user-facing stuff that needs to happen. And the caveat is that Google and Apple are talking to health authorities,

and they're talking to, quote-unquote, stakeholders, which means kind of anyone who is upset about this system potentially. So they're probably going to keep releasing information that gets more specific here. So at this point, we're doing a lot of speculation, and their job now is to convince people that they're doing a good job, and then release something in mid-May. Yeah, and they said today that they've been working on it for two weeks, I think it's notable that they let Apple take the lead in sort of announcing it.

I think they under the cover of Apple's privacy attitude, they're like obviously going with that versus Google's privacy attitude. But they're very insistent that this will remain private. To their credit, they've already released a spec for people to push on this. They're going to keep doing that. But as it happens, Casey, yesterday in your newsletter, you wrote exactly why these kinds of Bluetooth-based systems are somewhat controversial. and there's a lot of skepticism about that.

So walk me through that part of it. Yeah, so, I mean, a couple big things. The first one is adoption. Other countries around the world have implemented similar systems. And to the extent that they have been voluntary, most people have volunteered not to participate. So in Singapore, which is run by a fairly authoritarian government, they were able to get about 12% adoption.

And what that means statistically is that if you are walking around the neighborhood and another person is walking around the neighborhood, and one of us has COVID-19, and I don't know, we stand next to each other long enough for one of us to get infected. The odds that both of us have that app on our phone are about 1.44%. Now, it will be different in real-world conditions, right?

Maybe in San Francisco, a lot of us will download the app, and then, I don't know, like a Miami Beach, very few people will download the app, but getting people to adopt it is a huge challenge. The second major challenge is just the public health experts that I've talked to question the effectiveness of Bluetooth, particularly to identify these moments where the disease is likely to spread. My understanding of Bluetooth is that it can't always tell when you're within six feet of someone or not. It may pick up people who are much farther away from you, potentially even between walls, in which case you can imagine a lot of false positives. And so I still have a lot of questions about what happens when, you know, I get, you know, notification ordering me to quarantine because, you know, my next-door neighbor who is like eight feet away, but through two different walls, you know, might have pinged my phone and identified me as having been exposed.

Addie, do they talk about this at all? Yeah. So the idea is that they're kind of just promising that Bluetooth low energy has enough specificity and enough high-level features that they can kind of push on to make sure that it only grabs stuff within six feet. So say they can sort of figure out the strength of the signal and then gauge how far within range that means that the other phone is and that things that block the coronavirus will also block Bluetooth signal. It seems like a lot of how well this works is going to depend on the sensitivity and the flexibility of Bluetooth low energy here. And this kind of reason to another point that you specifically are talking about earlier as we were kind of going through the news. there's a part of this that feels somewhat easy for Apple to do. Apple controls their whole platform. They can push the alerts. They know what Bluetooth antennas are in their phone. They know the radio characteristics of their phone. But Android is an open platform. There are many, many different kinds of Android phones. Google historically is not great at pushing updates. And then there's a split in platform adoption between rich people and poor people. And that seems like it will have as much of an impact on this. As anything, like Casey, I think adoption is easy.

If every single phone in the United States running one of these two operating system gets an alert that says, do you want to turn this on? Presumably you're going to see more adoption. I mean, look, for God, six, 800,000 people downloaded Quibi this week. Come on. Well, okay, sure. But, Neil, like, let's keep in mind. Like, that argument would be more compelling to me if there was one national app and you had a coordinated federal response and the president of the United States was going on TV saying, please download this app. What instead we're going to get is that the North Dakota Department.

of Health and the San Francisco Department of Health and all these other departments of health are going to be developing their own app. And it is going to be up to all of them to drive adoption of their own app, which strikes me as just a massive waste of resources at a time when we need all of these resources. It's not clear to me that that's how the system's going to work. Like, I'm not saying it couldn't work that way, but it seems entirely plausible that one of the many agencies or like universities that has been trying to build visualization tools could get the blessing to just release something that has. has the official approval of the CDC or something. Again, this might not happen. Obviously, the dumbest possible thing tends to happen. But I don't see a reason why it would not be able to happen. So, is there any clarity on the sort of platform split? Because that's something that really occurs to me. Like, the people who are working in Amazon warehouses, the people who are working in the grocery stores, the people who are out in the world doing a lot of these essential jobs, you know, a lot of them do not make as much money as the people who are privileged enough to stay at home. And they might have Android phones.

just statistically that's how break down. If you have a five-year-old Android phone that doesn't have Bluetooth LE, how does this help you? How does this help everybody? You could make the really, really big-scale argument that anything that helps any small number of people helps everyone because this thing is so contagious. And the whole point is that we're just trying to get this under control at all. But it's a huge question.

First of all, the Amazon issue, I'm not even sure they have their phones while they're on the floor. So that raises its whole other set of questions. But yeah, a bunch of the places that have been hardest hit by this, especially in New York, are the poorest neighborhoods. And those are also the places that are going to have, a, sort of the hardest time getting tested in some ways. And this system depends a lot on testing. And they're, yeah, going to have old phones. It's going to be a little bit harder for them to get updates.

Google can push the updates in ways that aren't just the system-wide updates that take forever to get there. But even so, it's going to be a sort of extra barrier. Nicole, when you hear about all these challenges, if you take those away and say, okay, the best case scenarios, this works, the federal government wakes up. This is truly the best case scenario. But let's say all the correct things happen. How much do you think this helps and what are the next steps it needs to be effective? So, like, more contact tracing is good on a large scale because you want to be able to identify the people who potentially have been exposed to coronavirus, right? Like that's sort of the end-all be-all. Like you were saying before, this is extremely dependent on test. So contact tracing means nothing if we can identify the people who are sick and then identify if their contacts have gotten sick. So that's why the sort of public health message is test and trace, which is you have to test people and then you have to do contact tracing. You know, the benefit that this adds on top of if we just hired a million people to do regular gumshoe contact tracing is that it could potentially pick up people who you didn't know that you interacted. with, so someone who you were in a subway car with for a 45-minute subway ride or someone

who kind of followed around a grocery store for a long time or something like that, you potentially have a better way to pick up that sort of contacts that would maybe close some of these nets a little bit more closely. And, you know, experts talk a lot about how we need to do contact tracing better and we need to kind of bolster it in a really big way. That does not mean that technology replaces anything. It sort of augments the tools that we already have, which are very low tech and require investment and require people to do work. Can I see you in your newsletter, you brought up the idea that Bluetooth, there are like people Bluetooth cannot find. And you brought up the notion that

false positives would cause a lot of panic. Can you just walk through those two examples? Yeah, so there is a psychological cost to receiving a message telling you that you may have COVID-19, right? That's going to be a very scary moment for you. You're going to have to immediately seek out testing. And what people will tell you is, well, we want more false positives than false negatives because the consequence of a false negative could be that you get seriously ill and maybe die. At the same time, a system that is continuously falsely notifying, people that they have COVID-19 could undermine trust in the public health system at a time when

we need to be building that trust. So I just, that is something I will be paying very close attention to and we'll sort of see. And then the first point was just about people's struggles with Bluetooth generally. No, you have this line that in the piece from yesterday. It said Bluetooth cannot find this man. Yeah, okay, great point. So I talked to this guy, Tom Frieden, who used to run the CDC. And he tells me the story about when he was working for the New York City Health Department, they had this tough case where a young man had been diagnosed with multiple drug-resistant tuberculosis. He was in a juvenile detention center, and he escaped from that detention center. And they were very worried that he was going to infect other people.

And you cannot devise a Bluetooth app that finds this man. You have to put a team of people out on the street to go get it. And in this case, they did do that. But Tom's point was, if you really want to do the tracing part of testing and tracing, you need to hire human beings to do that. Human beings are very good at this, right? And this seems like a case where we might be so tempted by the shininess of the idea of this infinitely scalable Bluetooth app that we're all going to download and it's going to work magically that we turn our back on a project where we can hire up to 300,000 people to do this work, which is how many people Tom says we need. And there are 300,000 people out there who could use a job like this, right? A lot of it can be done over the phone.

So I don't want to make sure that as we keep talking about tech solutions, we don't lose side of the fact that there is an established human-based solution that we could also be investing in. Yeah, I mean, that's kind of the party line from a lot of public health experts when asked about technology to help with some of these problems. It's that the technology is good and useful and might help us a little bit, but it is not a replacement for the tired and true methods that we have been used. using and need to invest in. You see, I think California and Massachusetts are both have programs

where they're going to be hiring a bunch of contact tracers. The CDC says allegedly they promise they're going to be hiring more contact tracers. I mean, it depends how much you trust what the CDC is saying right now about what those plans will look like. But this is something that people are also talking about doing and, you know, it's less shiny, obviously. But it is equally, if not, probably a lot more important. I have to say the fact that we keep calling the shiny when it's Bluetooth, like there's like a deep, deep hilarious irony to be like,

Bluetooth will save us. And it's like, ah. This is fun. I remember when Bluetooth seemed really exotic when I was in high school. You know, I think it's interesting when we talk about this in the way that, you know, is anything better than nothing. It's sort of similar to the way we talk about masks right now, right? Like the types of cloth masks that we've all been told by the CDC that we should probably wear when we're out in public to try and prevent ourselves from spreading

any potential undiagnosed coronavirus around. Do not actually do a ton. They probably don't actually trap that many viral particles, but they're a potentially good way to maybe reduce transmission by 10% from each person, by blocking 10% of what they're spitting out of their mouths when they talk. And when we're in a situation like this, reducing transmission a little bit. is good as long as the risks of that don't outweigh that small amount of benefit. Also, I don't have to spend the weekend sewing an app.

Look, we are living in a really scary world where the federal government that is supposed to be protecting us has really dropped the ball. And so I think more than in other scenarios you can imagine, we actually do need to rely on those 10% solutions. And that's why I remain open to the idea of this Bluetooth-based tracking, even though the public health officials I've talked to are deeply skeptical of it. It may be the case where, you know what, for 90% of people who download it, it's not helpful, but it is helpful for a grocery worker or somebody who does have a long daily subway commute or somebody who's working in an Amazon warehouse right now, right? And if these apps can help them, then on balance, maybe it was worth doing.

Actually, Casey, I have a question about, since you've talked to actual experts about contact tracing, do they think that ethical questions aside, we would have something that's significantly better if we used, say, China's system of this incredibly invasive, you have to scan a QR code system. You know what's interesting about that QR system is my understanding is that the information that powers the algorithm, which we should back up a second and say what Adi is talking about. In some parts of China, in order to enter a building, you have to show your, personal health QR code. And my understanding is that it is unknown exactly what goes into the determination if you get a green, yellow, or red code, right? Green means go yellow and red bean,

go home. But it supposedly involves your self-reported answers to a questionnaire and where else your health code has been scanned. And what is the public health situation in those areas where you have been with your health code. That is, in my mind, actually not that much more invasive than some of the things that we have talked about. There are definitely, you know, invasive things that China has done, but the health QR codes might not be one of them. You know, I think the more invasive thing that China has done is that if you have symptoms, they send you to some sort of, you know, infirmary or quarantine area where you can be away from your family, not spread the disease to your family. And the experts I've talked to say, you know, we also need that solution here in the

United States. There are some people who cannot safely quarantine at home. Where are we going to put those people? That's another really big question that we can't solve with Bluetooth. Man, we're just running up against limitations of Bluetooth left and right. That is the big problem here. So, Casey, you're describing a, like, I would say a controversially invasive surveillance system. What this sort of Bluetooth idea is, right, it doesn't try. your location, but maybe these health apps that the governments are going to build will actually ask for your location because that's useful. When you do test positive, Nicole, like that information gets immediately sent to the government

anyway, right? Yeah. So, you mean, your health records exist and the health departments are tracking positive cases anyway. And I think it's important to note also that a normal contact tracing system with a human is also very invasive to your privacy. Like, this person is coming to you and saying, tell me every. everywhere you've been and who you've been with, and I'm going to ask you health information

about yourself, and then I'm going to go ask other people health information about themselves. So we're already, you know, public health ethics has a different bar than sort of other types of privacy ethics. That's something that exists already. Like we're already sort of working in a different framework than we would normally be thinking about things like your health privacy. So there are very important ethical principles involved. They're just different than the ones we might normally think about with health care. Can I say something about that? Because the most interesting thing to me about the Apple Google announcement today

is how much time they spend talking about privacy. And when you think about the past three and a half years, you understand why. We have just had almost four years of debate about the data privacy practices of these companies. And, you know, they've all made a lot of mistakes along the way. So it makes sense that they would invest so many resources in trying to make the most privacy forward version of this that they can and like to some extent good for them but they don't invest any energy at all in their announcement in talking about the efficacy of this solution right there is

everything else that these companies have done they have sent me a list of quotes from public health officials talking about how useful it's going to be to them right when google put out a a community mobility map they sent me like nine quotes from public health experts saying this thing is is great right and i've talked to public health experts who love the the mobility map that both facebook and google have done. And it is just so interesting to me that we're so focused on the privacy implications when, as you know, Nicole, privacy is not our number one concern here. And I understand that we are opening a Pandora's box if we don't take privacy seriously. And I'm not saying that these companies were wrong to take privacy seriously, but it is just crazy to me that privacy talk has like crowded out all discussion of

whether this stuff works. I don't actually know if, I think it is partly that, but I think some of it is maybe that we have taken for granted for so long that these companies know literally everything about us. And so, of course, they should be able to do this. Of course, it seems like the easiest thing in the world. Like, a lot of people have more or less written that in the last few weeks. And so I wonder if some of it is just that we're taking for granted based on the discussion of the last few years about how invasive all this technology is, that it can do a thing that, like you've been saying, maybe it can't. Yeah. Well, it's okay. I mean, I'll just ask you directly. Like the company that kind of is under the most fire for privacy issues is Facebook.

Facebook knows a lot about you. It knows a lot about your social graph. It knows a lot about where you've been. I mean, Facebook knows a lot of things. Could Facebook build something like this? So I've been asking Facebook about this. And I think there is a deep level of uncertainty there that this would work. Facebook has been eager to participate in the response.

They have these disease prevention maps, which show sort of anonymized aggregated data that health officials say is useful. They're also providing this form where you can go to a Carnegie Mellon site and self-report your symptoms, which can create a new kind of hotspot. Those are things that they feel really well equipped to do. I suspect it might have something to do with not owning the operating system that puts them in a tough position. Now, they do have an identity graph, right, which says.

seems really useful. But, you know, I was talking with someone there who was like, you know, Facebook has maybe 80% penetration in the United States. Google is apparently similar. So even at that level of adoption, you still wind up missing a lot of people. And so I think there's some reluctance there to make a bet on something that they don't think can reach everyone. Yeah. I think it's just really interesting to see where the technology has to live. And it's like, yep, you write the code that controls the Bluetooth radio. It's actually that level of depth into the hardware. You can't just distribute an app.

And I think it is remarkable. I was joking with Casey earlier. Like technology aside, vagaries of Bluetooth transmission, attenuation aside, all that stuff is really hard. Do you know what also is really hard? Getting Tim Cook and Sundar Pichai to tweet at the same time. Like, that's just like a dozen people had a week of meetings just to arrange that. Right?

And then you've got to solve all the tech problem. So it's remarkable they come together. I hope it's helpful. Nicole, what do you think, when do you think contact tracing becomes like a big part of the conversation? It seems like it hasn't been what we've been talking about yet. But as the conversation about when does this kind of come to an end, that seems like a big part of it, when do you think that that becomes a bigger part of the national conversation? Yeah, I mean, that depends a lot on when case numbers start to come down enough that it is, you know, feasible to do contact tracing for everyone who is. sick and also we need to raise to increase the number of people who are doing that contact tracing so that bar is is higher than it is right now. We're not doing contact tracing in New York City. Like we just, it's impossible because there are thousands of people getting sick every single day and that would require like astronomical resources. When you think about sort of the curve of the epidemic, conduct tracing is really important when cases are low at the beginning and can help

keep things low, which is when you see a country like Singapore that did that at the beginning and kept things low. And it's important when the curve starts to come down so we can try and keep it from pinging back up again. So I think that, you know, it's going to depend a lot on what that data looks like. I can't really give a time frame. I just know it's going to be when we ramp up our capacity to trace and when the numbers of cases are low enough that that capacity can handle it. And I think that does line up with their sort of stated timeline of when the APIs will be ready and when the operating system updates come out. That's still weeks to months away. Right.

Yeah. Addy, what happens next on the sort of like platform side of things? Is there any concrete next steps from these companies? It seems like it's just them reporting that they have talked to people who might criticize or work with them and they will release more information. Like the apps are supposed to launch in, or at least the API is supposed to come out in mid-May. So we're going to have to learn who's working with them, what the apps look like, what kind of level they're operating at geographically, how you're going to get results reported. Several of us have talked about how they're going to prevent trolls from just claiming they had COVID-19, which presumably they have a system for, but we don't really know much about it. Yeah.

So we just need to keep watching for explanations of many, many obvious questions that I'm sure they're thinking about, but I'm not sure they're. they've nailed down an answer for. I think the thing that they've done here that is smart and you would expect them to do is they know they need the sort of security community to beat up on this. And it feels like that's actually their first step is saying it out loud and then letting those folks kind of attack their plans and make sure they're buttoned up. Yeah. All right. Well, thank you all so much. I know this was an emergency podcast and I dragged you on, but I think it was really useful. Thank you to everybody for listening. We'll be back soon. We're going to be covering this very closely into next week in the weeks beyond. You can

And check out Nicole's explainer on contact tracing. Addie wrote the post today with Russell Brandem on what this thing actually is. Also a post sort of at high level on how you might be able to get it, which I think is important. Check out Casey's newsletter. Casey, do you want to plug your newsletter? Yeah. I mean, I think of it as our newsletter, Neelai, but it's over at theverge.com slash interface. Excellent.

All right. Thank you guys so much. We'll talk to you soon.