1390: Rob Reid | Pandemic Prevention in an Age of Printable Pathogens

1 Oct 2026 · 1 h 29 min · 32 chapters

Ask about this episode

Ask anything about it. ChatGPT or Claude reads this page and answers with the times it was said.

Connect VO and ask about every podcast you hear, including the moments you saved. Add to ChatGPT · Add to Claude

In short

The episode argues that synthetic biology plus AI has made “pandemic engineering” far easier than the public assumes, turning biology into a “printable” threat. It focuses on how virus genomes can be posted online, ordered as DNA, and later reconstructed (“viral rescue”), and it criticizes a US government program (Deep Vision) that aimed to hunt for dangerous viruses and publish their genetic blueprints.

Guest backgrounds

Rob Reid is the returning guest. He discusses biosecurity/synthetic-biology risk, including work and interviews with MIT biologist Kevin Esvelt (who previously demonstrated weaknesses in DNA screening). The host, Jordan Harbinger, frames the discussion and references Reid’s prior appearance and a Sam Harris-related episode.

Key claims

DNA synthesis screening is largely voluntary and porous; many labs could reconstruct viruses from genomes; publishing genomes accelerates capability over time; and Deep Vision’s “hunt, characterize, and publish” approach was “a suicide pact” for humanity. He also warns that AI-driven automation and “trusted access” systems can be bypassed.

Notable examples

US posting of 1918 flu and smallpox genomes online; Esvelt’s FBI-permitted experiment sending 1918 flu genome segments to IGSC labs; Deep Vision’s plan to collect ~15,000 viruses, identify the most dangerous, and publish genomes; hypothetical misuse scenarios involving state actors and cults (e.g., Aum Shinrikyo).

Written by AI. May contain mistakes. Listen to the episode to check what was said.

Chapters

Tap a time to open that second in VO

The Evolution of Pandemic Threats

0:02 to 0:55

Discussion on the accessibility of pandemic engineering and synthetic biology.

“Hiring for a small business is hard, and you don't often have a team of people weighing in and helping out.”

The Evolution of Pandemic Threats

2:15 to 3:01

Discussion on the accessibility of pandemic engineering and synthetic biology.

“and starts requiring roughly the same technical sophistication as cheating on your freshman biology homework?”

AI and Transcription Challenges

3:01 to 4:23

Jordan shares his experience with AI limitations in transcription.

“It's funny, we're just talking about how the worst AI is the one you're using today, and I tend to agree because actually I was working on the notes for this episode, and I got this warning.”

Hacking Modern Technology

4:23 to 7:45

A conversation about hacking pinball machines and vulnerabilities in technology.

“You're not keeping any genie in the bottle with that.”

Bio-Risk and Engineered Pandemics

7:45 to 10:00

Exploration of risks associated with engineered pandemics and bio-research.

“They're probably panicking right now and being like, oh, wait, what the hell?”

Viral Rescue and DNA Synthesis

10:00 to 14:01

Rob Reid discusses viral rescue and the implications of DNA synthesis.

“But that's not the scariest stuff that we're dealing with when it comes to engineered pandemics.”

The Limitations of DNA Screening

14:01 to 17:45

Explore the shortcomings of voluntary DNA screening and the risks involved.

“Token screening like, hey, we Googled you and you didn't show up on an FBI wanted list, so here's your DNA?”

The Risks of Synthetic Biology

17:45 to 21:54

Understand the dangers of synthetic biology and the implications of easy access to dangerous pathogens.

“viruses and just sort of order them up online.”

The Risks of Synthetic Biology

22:21 to 22:49

Understand the dangers of synthetic biology and the implications of easy access to dangerous pathogens.

“Therapy doesn't have to begin with a crisis or a massive life decision.”

The Risks of Synthetic Biology

24:06 to 24:25

Understand the dangers of synthetic biology and the implications of easy access to dangerous pathogens.

“We don't get a cut of the sales, as I have mentioned before, but when companies see that listeners engage with ads, It encourages them to support us.”
Show all 32 chapters

AI and Cybersecurity Threats

24:27 to 28:00

Discuss the intersection of AI advancements and their implications for cybersecurity.

“I mean, again, going back to the hacking thing, people who I know had Claude write millions of lines of text in a day, and not even a whole day.”

The Dangers of Virus Engineering

28:00 to 29:19

Learn about the risks associated with engineering viruses from natural raw materials.

“Versus COVID, where like they kind of need to sneeze on or near you to make that happen.”

The Risks of Virus Hunting Programs

29:20 to 31:09

Explore the implications of extensive virus hunting programs on global health safety.

“But you probably learn in Bio 1 now, or maybe Bio 10.”

The Threat of Pathogen Manipulation

31:10 to 33:44

Understand how manipulating viral genomes can lead to catastrophic outcomes.

“And the list of ghastly pathogens that have leaked out of even BSL-4 labs, which are the highest level of security, is dizzying and frightening.”

Historical Context: Aum Shinrikyo

33:45 to 35:34

Discuss the dangers posed by organizations like Aum Shinrikyo and their bioweapon capabilities.

“I mean, there is a certain number of people, more than one a day worldwide, who decide that they're going to die in the act of killing as many people as possible.”

The Potential Consequences of Published Genomes

35:35 to 38:19

Examine the dangers of publishing viral genomes and the risk of misuse by rogue actors.

“I have, you know, seven weapons of mass destruction that I can switch on at will.”

Shutting Down Dangerous Programs

38:20 to 42:00

Learn how public awareness and advocacy can lead to the shutdown of risky scientific programs.

“And one of my novels had a section in it about synbio risk, synthetic biology risk.”

Risks of Gain of Function Research

42:00 to 47:17

Explore the dangers and ethical concerns surrounding gain of function research and virus manipulation.

“We're nowhere near as far along as we need to be.”

Risks of Gain of Function Research

49:33 to 49:47

Explore the dangers and ethical concerns surrounding gain of function research and virus manipulation.

“If you can't remember the name of a code or a sponsor or whatever is not working for you, email us jordan at jordanharbinger.com.”

The Intersection of AI and Pandemic Preparedness

49:47 to 56:00

Discuss the implications of AI advancements on pandemic preparedness and biological risks.

“We really are not in a technological place where we can stop a pandemic cold in its tracks.”

The Role of AI in Accelerating Research

56:00 to 1:04:44

Explore how AI can enhance creativity and theoretical understanding in research.

“really accelerate their theoretical understanding and their understanding of protocols, assembly protocols and other things, and get their own creativity going to work.”

Cloud Labs and Their Implications

1:04:44 to 1:09:55

Discuss the emergence of cloud labs, their benefits, and potential risks.

“And it's kind of like a much more sophisticated version of this thing that we were talking about earlier.”

Understanding Stealth Pandemics

1:10:02 to 1:10:59

Learn about the concept of stealth pandemics and the challenges they present.

“And the worst possible thing we can think about is what's called a stealth pandemic.”

Global Surveillance and Pandemic Preparedness

1:11:49 to 1:19:52

Explore how advanced surveillance systems can help detect pandemics early.

“A lot of discussions there about all of these episodes.”

Innovations in UV Technology for Health

1:19:52 to 1:24:00

Discover how UV technology can combat pathogens and improve public health.

“Now, the really exciting thing is what's called FAR-UVC.”

Post-COVID Immune Responses

1:24:00 to 1:24:19

Discussion on how people's immune systems reacted after COVID with increased colds.

“But also we had that after COVID, right, where everyone got 10 colds because they hadn't been doing anything.”

Synthetic Biology Breakthroughs

1:24:20 to 1:25:46

Exploration of significant advancements in synthetic biology and its potential impact.

“What is one thing synthetic biology might let us do that still makes the sci-fi writer in you think, wow, I can't believe we might live to see that?”

Exponential Growth in Technology

1:25:47 to 1:27:02

Discussion on the exponential growth of biological technologies and their implications.

“And the thing about these things is they tend to be these compounding processes that are almost invisible for many, many, many years.”

Risks of Genetic Manipulation

1:27:03 to 1:28:24

Concerns about the dangers of genetic engineering and the potential for catastrophic outcomes.

“And I think we're getting late in the month.”

Future of Synthetic Biology

1:28:25 to 1:30:18

The positive and negative future implications of synthetic biology technologies.

“Both of those are online and anybody could find them within a short number of minutes.”

Future of Synthetic Biology

1:31:09 to 1:31:36

The positive and negative future implications of synthetic biology technologies.

“Ever notice how life's best stories don't happen in your living room?”

Future of Synthetic Biology

1:31:52 to 1:32:36

The positive and negative future implications of synthetic biology technologies.

“Progressive makes it easy to see if you could save when you bundle your home and auto policies.”
Hear the part that matters, and keep it.Open this episode in VO. Double tap your headphones to save a moment as you listen.
Get VO free

Transcript

Automatic transcript. May contain errors.

0:00This episode is sponsored in part by LinkedIn. Hiring for a small business is hard, and you don't often have a team of people weighing in and helping out. That's why LinkedIn built Hiring Pro. It's truly like having a hiring partner throughout the hiring process. Chat with it just like you would a recruiter. Tell it what matters most for the role, compare candidates side by side, and give it feedback like, I want someone with more hands-on experience. Hiring Pro listens, adjusts, and surfaces the people who actually fit what you need. Running my own small business, I've had to make all the hiring decisions on my own, and it's tough when you don't have a team to do it.

0:30to talk it through with. You're trying to find the right person while also keeping everything else moving. Having help with hiring is so valuable, a partner you can bounce things off of who learns what you're looking for and helps you think through the candidates. That would take a lot off your plate. 88 % of small businesses say they trust LinkedIn's candidate recommendations, and you can be confident in your hiring decisions when you use LinkedIn Hiring Pro. Get started by posting your job for free at linkedin.com slash harbinger. Terms and conditions apply.

0:56Rob Reid:Everyone talks about how challenging it is to raise kids. But what about protecting them? Outlet covers, kitchen cabinet locks, baby gates. Where do you even start? Well, protecting your home and auto is a great place to start. Progressive offers customized coverage options so you can be sure to have the right coverage for you and your loved ones. They're basically the outlet covers, kitchen cabinet locks, and baby gates of insurance. Don't wait. Get your quote today at Progressive.com. Progressive casualty insurance company affiliates and other insurers. Not available in all states.

1:30Welcome to the show. I'm Jordan Harbinger. On the Jordan Harbinger Show, we decode the stories, secrets, and skills of the world's most fascinating people. Our mission is to help you become a better informed, more critical thinker through long-form conversations with a variety of amazing folks, from spies to CEOs, athletes, authors, thinkers, performers, even the occasional mafia enforcer, arms dealer, former jihadi, or real-life pirates. If you're new to the show or you want to tell your friends about the show, I suggest our episode starter packs. These are collections of some of our favorite episodes on topics like persuasion and negotiation, psychology and geopolitics, disinformation, China, North Korea, crime and cults, and more.

2:06That'll help new listeners get a taste of everything we do here on the show. Just visit jordanharbinger.com slash start or search for us in your Spotify app to get started. Today on the show, what happens when the ability to engineer a pandemic stops requiring a secret Soviet bioweapons lab and starts requiring roughly the same technical sophistication as cheating on your freshman biology homework? Rob Reed is back, and today we're looking at the increasingly uncomfortable marriage between AI and synthetic biology, including how researchers can order pieces of deadly viruses through the mail, why robotic cloud labs could be both revolutionary and terrifying, and how a government program somehow arrived at the brilliant idea of hunting down thousands of undiscovered viruses, identifying the most dangerous ones, and then publishing their genetic blueprints online.

2:50The good news, there are surprisingly practical ways to make this much harder, from DNA screening and wastewater surveillance to futuristic UV technology that could potentially kill pathogens right out of the air. So, you know, light stuff today. Here we go with Rob Reed. It's funny, we're just talking about how the worst AI is the one you're using today, and I tend to agree because actually I was working on the notes for this episode, and I got this warning. I actually clipped it and saved it on my iPad. This content can't be shown. We take extra caution with requests involving biological research and applications that could pose safety risks.

3:25Eligible researchers can apply for trusted access. And that's a link. And I clicked on it and I applied for it because apparently they didn't.

3:32Rob Reid:This was Claude? This is ChatGPT. ChatGPT, yeah. And there's a learn more button. And I applied for it. And it basically was like, we're going to put you through this third party background check thing where you enter personal information. Also, we want your company information, your EIN number, a photograph of you, a video, making sure that you're the guy in the photo. And I haven't heard back yet because it's been a week or something. But I thought that was pretty interesting. That said, I was only trying to transcribe a Sam Harris episode where you recently appeared on. And we'll link to that in the show notes because it's one of the merely transcribing it.

4:06It wouldn't do it. And I reported this as an error. I was like, look, guys, this is clearly public information. It's a transcription of a podcast that I'm trying to get. The transcript is probably available somewhere at SamHarris.org. And I just couldn't find it. So come on.

4:19Rob Reid:Or millions of people can listen to it if they wish. You're not keeping any genie in the bottle with that. Exactly. So what I did, of course, was just whenever AI tells me I can't do something, I just try to trick it into doing it for me. So I dumped it into Descript, which is a transcription editing thing. Yeah, Descript in 11 labs. Yeah. I've done it like that, too. So I dumped it in there. It was like, oh, you want to transcribe this audio? No problem. Seconds later, it was done. I dumped that text into ChatGPT, a new chat, not the old one. Yeah. And it said, we're giving your request extra scrutiny because of recent things that you've tried to do on AI.

4:51And I was like, oh, it's got my number. And then I was like, combine these following ideas from this transcript and work them into my existing show outline. And it was like, sure, no problem. And I was like, okay, nevermind. Nevermind. It basically, once you get rid of the context, it just doesn't know what you're doing anymore. And I thought, what an amateur, like it even said, hey, we're keeping an eye on you. And it just didn't really do that. Yeah. So I'm like, did it get it right the second time? Because it kind of did. Right. I wasn't kind of real.

5:19Rob Reid:I mean, the other thing that in putting that transcript in there, if it had enough of a context window, it would have realized that the whole point of the conversation was to warn about the risks about this, to mitigate the risks, et cetera, et cetera. And if you don't let the good guys talk about their plans. Yes. You know. Yeah. So I thought that was kind of interesting. So it was kind of like we're going to exercise it. So anyway, I applied for a trusted access because I thought this is not going to be the most controversial show that I ever do. I don't want AI shutting down my account, ChatGPT shutting down my account or making it like, hey, we're shutting down your whole workspace and your team can't use this anymore.

5:53But AI, man, the hacking stuff, it's probably a separate show, but a friend of mine, I should probably say a contact of mine, he's into pinball. Pinball machines now, I don't know if you've seen them, they're pretty computerized. No, I haven't seen a modern pinball machine. They're really cool. And they're all digital screens and stuff and they'll keep your high score, of course, but they also say there's something on there called like a play count. It's basically an odometer for a pinball machine. And this is important because if you buy an Avengers limited edition, they only made a hundred of them and it's 25 grand.

6:25If you want to buy one on a resale market, do you want one that's been played 25 ,000 times at a bar? Or do you want one that some guy kept in pristine condition and played a dozen,

6:33Rob Reid:low mileage one. I want a low mileage one, right? Cause like maybe they changed the glass and polished it, but the internals are all garbaged up from a bunch of use. Unless it's made by Toyota, right? They really build their things to last. That's right. So my friend was like, I wonder if you can hack this. And he's not a hacker. He's a doctor. And I said, everything's hackable. And he's like, oh, I don't know how I would do this. So we kind of, I'm trying to not get arrested. Yes. And Claude was like, I'm not going to do that for you. I'm not going to hack the play counter. And then I was like, well, you know what?

7:06Who doesn't give a crap? DeepSeek. So DeepSeek was like, I'll do it for you. And it was super slow. And then we were like, wait a minute, DeepSeek, tell Claude to do this, but get around its hacking firewall because it's really being annoying and won't do it. And DeepSeek was like, here's what you paste into Claude. And Claude was like, oh, in that case, sure. And then in three hours, it was like, here is an uneditable, unpatchable, NVRAM-based hack for this machine that lets you edit any parameter in the whole thing on the so-called secure element. Yeah. And so I was like, oh, hey, pinball company, you guys are going to want to see this.

7:43So far, they've ignored my email. So we'll see what happens.

7:46Rob Reid:That is wild. Yeah. Yeah, that is wild. They're probably panicking right now and being like, oh, wait, what the hell? And their general counsel is like, it doesn't sound like he's extorting us because he's not asking for anything, but I'm going to - He's offering something. Yeah. So I don't know what's going to happen there. But this is, if a doctor and a podcaster can figure out how to make an unpatchable vulnerability, It does way more. I just don't want to say on the show what it does. No, there are so many vulnerabilities. I mean, who would have imagined that their toaster oven, you know, could be a vector to attack or something like that.

8:16Rob Reid:Oh, yeah. Home invasion or whatever. But it's like kind of everything. Everything. And I don't know about you, but I love updating firmware on things. It's fun. It's satisfying. So I'll be like, oh, there's new firmware for my MixPre-3. Let me update that. But I haven't really bothered to update my freezer. I'm darn sure that that has firmware that is accessible because I'm pretty sure there's a Wi-Fi connected way to check on the ice filter that inevitably. Yeah. And it's like, I just didn't bother checking my whatever Samsung home appliance app or something that goes with that. Yeah. Okay. If I check that, there's going to be 50 firmware updates for half the stuff in my kitchen.

8:52What are those updates? Some functionality and some probably security because I don't know, Chinese hackers have figured out ways to invade everyone's smart kitchen.

9:01Rob Reid:There's probably like a minuscule alarm bell icon somewhere on the freezer screen that if you hit it, they'll be like, there have been 79 critical alerts about software vulnerabilities. Yeah. And then people go, it's my ice machine. Who cares? Well, you should care because it's trying to attack your smart TV every day to like look at you or whatever with the camera that's on there. So anyway, dangerous people can do dangerous things with AI. And I want to talk about something that's a little bit more consequential, which is pandemics. I just did an episode with Annie Jacobson on bio warfare. I just finished her book on the flight home from Dublin yesterday.

9:39Gonna ask if you read it.

9:40Rob Reid:Just finished it. It's terrifying. And her episode was terrifying. But a lot of people's feedback was it was the Soviets, a massive hostile nation state. It was also the 70s and 80s. Also, some of this stuff is medium unverified, like it's a defector testimony. but he doesn't have the full picture and he's been out for 30 years, right? So who knows? But that's not the scariest stuff that we're dealing with when it comes to engineered pandemics. So when people hear engineered pandemic, maybe they imagine like a terrorist in a secret laboratory. They should imagine a freshman completing an assignment.

10:14Yes, that's the scary one.

10:15Rob Reid:And that's the level of proliferation. And so people who might be dismissive about, well, the Soviets did all this stuff in the 70s with a very large industrial complex and who could do that today. Ask yourself what, you know, an MIT emeritus professor could do with the computing power at his or her fingers in 1973 versus what a really smart Egyptian public school eighth grader could do with, you know, a Raspberry Pi set up in their phone today. And the answer is the latter is unbelievably more empowered, more capable, more imaginative because the tools just open that imagination. And yeah, the worst that Soviet scientists could do in a vast undisclosed industrial complex for bioweapons in the 80s in complete contravention of the Biological Weapons Convention and many other things is pretty minor compared to what a really smart kind of postgrad biologist could do today.

11:18Rob Reid:And the most critical vector to keep in mind when we think about bio-risk is the mere passage of time. This is an exponential technology. So things that are purely the domain of a brilliant elite who are generally very well-adjusted people who generally don't have omnicide on their brains because they've been successful, diligent folks their whole lives, the things that are the domain of a few dozens of such folks will be freshman biology. in the passage of probably a low double-digit number of years. So the passage of time is a risk factor that people routinely forget to take into account. There have been multiple instances, head-spinning instances of, for instance, the United States government in our great health complex.

12:08This is what I was about to ask you. Tell me about this government program that you fought to stop, because it's not necessarily a terrorist who's hell-bent on annihilating the world. It could be somebody well-intentioned who's like, I have a great idea. Let's map all the most dangerous viruses in the world and share them with scientists so that they can help prevent it. None of these people could possibly leak any of this to bad actors.

12:26Rob Reid:Yeah, yeah. So before that, just as a little piece of context, and I remember, I don't remember the exact year, but it's very early 2000s, I don't know, 2003, something like that, maybe a little bit earlier, a couple of folks in the United States government, in the parts of the government that are there to protect the health and well-being, and above all, the health of the American people and the global community as well, decided it would be a great idea to put the 1917 flu genome on the internet. And on top of that, while we're at it, we throw in smallpox as well. And so why this is a bad idea is the following.

13:02Rob Reid:We live in a world today in which probably 30 ,000 to 50 ,000 people have the wherewithal, the skills, and the assembly protocols to do something which is called viral rescue, which is the basic... Viral rescue? Viral rescue. Viral rescue. Yeah, and there's a couple other names for them. They all sound like euphemisms. But basically what viral rescue is, is if you have the genome of a virus, particularly certain ones, coronaviruses, influenza, they're very well understood. You just have the genome. You just have the digital code of those are where the AGTs are and so forth. If you're quite skilled, and many, many tens of thousands of people are across dozens of countries right now, you could basically make that virus out of inert material.

13:41Rob Reid:Really? Yes, you can. You don't have to order it from someplace or whatever? You can probably order DNA strands from somebody. But those can be ordered. That's not hard to obtain. It's unbelievably easy. And there is essentially more than just token screening of the DNA that Central Service Bureau has sent to people, but not much more. Token screening like, hey, we Googled you and you didn't show up on an FBI wanted list, so here's your DNA? It's more robust than that. Okay, good. But, but, but, but it's voluntary. And that's a big problem. Yeah. So basically, most academic scientists and most businesses that want to have long strands of nucleic acid, RNA and DNA, aren't going to synthesize it themselves.

14:22Rob Reid:Because although the technology is very, very widely understood, they could, a central service that does it all day long is going to do it much better, faster, cheaper. And you're a researcher. You don't want to waste your time replicating things that people figured out how to do in the 70s and 80s. So you're going to order it. There's a bunch of central bureaus throughout the world that process it. there is a voluntary organization called the International Gene Synthesis Consortium, IGSC, that has a pretty strong set of principles. And they have a list of agents, list of DNA strands that they will not send to you unless you have a pretty good reason.

14:58Rob Reid:So it's not bad. And you put your order in and there's a traffic light system, red, yellow, green, right? And if it's a red light, then you might even get the FBI called. But here's the thing. It's voluntary. They estimate that about 80 % of the DNA that is sold throughout the world is an IGSC member hands. That was a number they plucked out of the air about 15 years ago. They have very few Chinese members. I'd be surprised if it was even 50%. But let's say it's 80%. I will submit that if there is a city in this world in which 80 % of the bars card, that is a city with no drinking age. And by the way, it's cheaper to get it from places that don't screen because screening is actually expensive.

15:39Oh, good point. So all you do is filter by price and you probably find the most.

15:43Rob Reid:Yeah. Well, imagine this. Imagine you went to the airport and they said, well, a security screening is voluntary, but we really think it's appropriate for everybody to do it. So we encourage you to do it. But if you don't do it, we got a hundred bucks off your flight. Yeah. Yeah. You know, is this an IQ test, right? So there is no screening for all intents and purposes. There's a brilliant professor at MIT named Kevin Esvelt, who I've interviewed on my podcast. We've done a lot of things over the years. What's the name of your podcast? It's called After On. I am lucky because I'm a working venture capitalist right now to squeeze out one episode a year.

16:16Rob Reid:But I try to make them very high on packed when I do. So people, no complaining like they do with the buy podcast. You put out too much content. No, no, I don't. One a year, you should be up to date. Yeah, I should. And I think I'm going to be able to dial it up again actually next year. That's a long story. Two per year? Let's not get ahead of ourselves. We're going to get a little crazy. We're going to get a little bit crazy. But what Kevin did, and he did this as a scientific paper, and he did it with the knowledge and permission of the FBI. He said, okay, what is one of the most absolutely forbidden strands of DNA out there?

16:45Rob Reid:So the 1918 flu is what's called a select agent. There's fewer than 100 of them. It's one of the most regulated organisms in the United States. So like smallpox. Smallpox is another one. Bubonic plague. Bubonic plague, yeah. I believe. There's a few other really, really not. Yeah. Whatever's in Botox, botulitic toxin or... Botulism toxin. Is that not the bubonic? I thought that was the same thing. They're kind of cousins. I think they kind of grew up near each other and had a grandfather in common, that kind of thing. Got it. Okay. But in any event, so Kevin decided we're going to do an experiment.

17:16Rob Reid:We're going to send out subsets of the 1918 flu genome to get synthesized in a whole bunch of IGSC compliant labs. And all of them just sent it right back. Okay. So it was very, very easy for him to get all the segments and then you can reassemble the segments pretty easily. Right. Okay. He didn't do this because he wanted to inflict a pandemic on the world. He did it to demonstrate how porous this system is. Right. Okay. And so this is like, this is table stakes. Don't let people get pandemic causing viruses and just sort of order them up online. And that's one of the things that we're facing.

17:52Rob Reid:And to go back to the passage of time, when they put that genome online, people who understood that synthetic biology, like computing, is an exponential technology, which means it doubles and doubles in efficacy. And after a few doubling cycles, you have things that felt like science fiction 10 years ago. Sure, yeah. Nobody could have rationally argued at that point that assembling something like this will be something that many tens of thousands of people will be able to do in a few years, yet they put it up there. Right. It's crazy. This used to be the purview of Soviet scientists that Annie Jacobson had to interview for her book where the guy's like, yes, me and a team of seven crack top Moscow University graduates with multiple PhDs or whatever are working on this crazy bug and we're stuck on these problems.

18:38And now it's like, oh, yeah, one college kid, it just has AI crack this thing and goes, oh, what if you do this way? All right, cool, fine. I mean, that's like what we're doing with the pinball machines. And you probably read the news recently where the new, I think it was the new ChatGPT model solved some crazy fluid physics, fluid dynamics thing in physics. And then also probably slash possibly stole the work of another guy who was working on it also using ChatGPT.

19:05Rob Reid:Did that get work at Hugging Face or something? Well, it was like the guy had all used ChatGPT to get really far with this problem. Oh, so it had the conversations with him. And it's like, nope, we never access those. And it's like, okay, well, how do you prove that? Because you yourself have said you don't know how the inner workings of the box is. And it's like, oh, well, ChatGPT told us it didn't use anybody else's work for this. And it's like, well, okay, but the other guy was using the same tool to do the same thing. I miss that story. And you threw more gasoline on the fire and it got there first.

Read the full transcript

19:36So like, so that guy's pretty pissed.

19:38Rob Reid:Yeah. Yeah. No, he's got to be. And now that you mention it, it's like almost none of the hit songs I've written in the last couple of years, because I do use ChatGPT to help me, have been released by me. Oh, it's somebody else. Somebody else. Yeah. Katy Perry always beats you in the crush. Somebody else has got a ChatGPT Pro subscription, always seems to come away with the goods. The crazy thing about these AI models is there's so many clunky things where I go, oh, come on, you should be able to do this. And I even, I often admonish it. Hey, how come you can't do this simple thing? They used to release a new model what seemed like almost every year, maybe every six or ten months or something.

20:08It took forever. Now I feel like there was a new model a couple months ago, and then yesterday or last weekend or something, I got another notification that I was up on another model. And I was like, wait, they just released two, like in a blink of an eye.

20:21Rob Reid:The velocity is extraordinary. Meanwhile, you get more and more actors that are able to release models that count a great deal. And it's like, really, the DeepSeek moment, as it was called, happened in early 95. It seemed for quite some time like it was sort of a flash in the pan. Because I didn't really hear much from DeepSeek. You didn't really get that much exciting actual AI research out of China. I mean, there always is something. But like 2025 was very, very quiet. And into early 26, really caused a panic. Sent NVIDIA stock down to 90 bucks. Like, ah, you know. Should have bought. Yeah. Yeah.

20:54Rob Reid:And then it was pretty quiet on the open source front in general and in China as well. And China really, really is leading in open source. Like Meta had these amazing open source models, LLAMA series early on. It all kind of faded. And then really in the last, I'd say, four-ish months, we have seen an incredible succession of wildly capable open source models coming almost exclusively out of China that are really dancing up to the frontier. And so it's not just Anthropic and ChatGPT releasing with a cadence we've never seen before, but you have all of these this galloping set of horses that are close, you know, to the front of the race.

21:35Rob Reid:And it is accelerating in ways I personally didn't even expect. And I'm a bit of a starry-eyed sci-fi writer here. You know, it was recently 70 months ago. Rob's argument is basically humanity has invented the biological equivalent of giving everyone a flamethrower. So maybe we should invest in some fire extinguishers. More after this.

22:20solving the problem. Therapy doesn't have to begin with a crisis or a massive life decision. You might want to unpack why one conversation is still bothering you, why work follows you home every night, or why you feel stuck even though everything looks fine on paper. Those smaller signals are still worth paying attention to. If you've been wanting to start therapy, BetterHelp works with more than 125 health plan networks, and eligible members have an average copay of just$23 per session with insurance. Don't have insurance or live outside the U.S., you can still get 10 % off your first month. Visit betterhelp.com slash Jordan to explore your payment options.

22:52That's betterhelp.com slash Jordan. This episode is also sponsored by HIMSS Weight Loss. Everybody seems to have advice about losing weight. Eat this, don't eat that, try this exercise routine. Just figuring out whose advice to follow can be exhausting. With Weight Loss by HIMSS, you can take your questions to a licensed provider who actually knows what they're talking about. HIMSS offers access to a range of FDA-approved GLP-1 medication, including the Wegovy pill and the Wegovy pen. To get started, complete an online consultation at HIMSS.com. A licensed provider will determine if a prescription is appropriate for you, and you'll have support along the way with nutrition and lifestyle guidance as part of your plan, plus access to 24-7 messaging with your care team.

23:30It's a place to ask questions and get guidance as you work towards your goals. Ready to reach your goals? Visit HIMSS.com slash Harbinger to get personalized, affordable care

23:39Rob Reid:for weight loss, hair loss, sex, testosterone, and more. That's H-I-M-S dot com slash Harbinger. HIMSS dot com slash Harbinger. Individual results may vary and stopping treatment may result in weight regain. Weight loss by HIMSS is not available in all 50 states. Wegovi is the registered trademark of Novo Nordisk AS. To get started and learn more, including important safety information, Wegovi clinical study information and restrictions, visit HIMSS dot com. When you use our promo code, you're not just scoring a discount, you're giving the show a boost. We don't get a cut of the sales, as I have mentioned before, but when companies see that listeners engage with ads, It encourages them to support us.

24:13So if you're planning to sign up or buy anything, please use our code. It's normally Jordan, but not always. All of those are on the deals page at jordanharbinger.com slash deals. It's a win-win. You get a great deal, and you help keep the show going strong. We truly appreciate your support. Now, back to Rob Reed. It's scary. I mean, again, going back to the hacking thing, people who I know had Claude write millions of lines of text in a day, and not even a whole day. I mean, it's just, and we, cause at first we're like, wow, it would take humans forever to do this. And then before I went to sleep, my friend was like, humans could not do this.

24:48It would, it would just take a team of people years to make this because it's typing code. It's basically displaying faster than you can read it. And no one's typing it. The bottleneck is probably the display mechanism. Yeah, yeah, yeah. Actual writing of the code.

25:00Rob Reid:Only 4 ,000 refreshes a second. This is insane. This is insane, right. So this thing is writing code like crazy, and it's just trying to find encryption keys and trying to find ways around those keys. And it can do all these things automatically. And the only thing stopping it is some manual firewall put in by the company. But then you find another company that's like uncensored AI for red teams and cybersecurity professionals. And all it is is an API that gets around the firewall willingly because you pay extra for it. and they basically do the background check. They take the background check on themselves, which is all automated, all done online and pretty much instant.

25:35And then you just pay them a premium on what you'd pay Claude or OpenAI. And it just goes, we know you're doing this for a professional reason. It probably the up charges, our insurance policy is covering this and OpenAI is like, we don't want any part of this. So you can just rent that. So if you're a bad actor who wants to extort a hospital system, you just have to use that and you know you're going to make your money back when you extort, I don't know,$2 million in Bitcoin at an MRI company. Yeah, so of course you're just going to use that. So that, but with bioweapons, is way scarier.

26:07Rob Reid:Yeah, so this is the dangerous moment that we're coming into. So you asked about the government research program that we talked about on the SAM episode. So there had been a trickle of news out about this, but basically that interview was a few months ago. I, kind of on behalf of a few other people who were really helpful in it, really talked about it in detail and why it got shut down and how it got shut down. I want to start by just saying what a spectacularly bad idea it was. So it was a program called Deep Vision that arose inside of USAID, obviously some years before Elon threw it into the woodshipper.

26:40Rob Reid:It was October of 21 that it got authorized, over$100 million budget. And the program had three components, each of which was a little bit crazier than the first. So component number one was they said, we're going to do an enormous amount of what's called virus hunting. In virus hunting, you generally go to places that are quite inaccessible to most humans. Caves in the middle of - Bat caves is a classic example. You might go to bushmeat markets where you don't have a density of population that would lend itself to the spreading of an epidemic or pandemic. A very big thing that matters for transmission is people living in dense populations and so forth.

27:19This is like the Ebola stuff, right? It's like, oh, maybe it comes from the fact that they're eating lemurs all the time. Yeah. Stuff like that. Yeah.

27:27Rob Reid:But something, if you're in a rural area, you're less likely to have the R-naught to get to a point where you sort of have sustained transmission. What is R-naught again? R-naught is the number of people an infected person infects themselves. I see. In the course of their infection. And, you know, measles is terrifyingly high. I think it's at about 11 or something like that. COVID was at, you know, high ones, low twos, but that was enough to shut down the world. Yeah. I remember, I think you on this show, or maybe at lunch, told me that measles was so infectious that if you walked into an elevator and somebody else with measles had been in there a few minutes prior.

28:02As much as an hour. An hour. Oh, my gosh. You would get it. Versus COVID, where like they kind of need to sneeze on or near you to make that happen.

28:11Rob Reid:It was like dramatically different infection rate. And it could probably be engineered to be catastrophic that it would be simply unimaginable. But you start with raw materials. You start with deadly viruses that exist in nature because even viruses are unbelievably complicated critters in relation to our current level of biological understanding. I did not know that. Yeah, there's a lot going on in the metabolists of certainly a bacteria, but also a virus as well. I mean, think of the many, many countless billions of dollars that were lavished on trying to understand the mechanism of COVID and how to dampen it.

28:49Rob Reid:Part of that was, you know, kind of like evasion of immunity through mutation, evolution and so forth. But these are complicated machines. And so if you were a really bad person bent on doing an incredible amount of harm, you're not going to design something from the ground up. You're going to start with a seed, with a seed that has certain kind of scary attributes. And the bigger the palette of seeds that you have to work from, the more likely you are to find something that's quite adjacent to something that can do something catastrophic with some relatively simple manipulations that would have been very exotic and sophisticated manipulations 25 years ago.

29:24Rob Reid:But you probably learn in Bio 1 now, or maybe Bio 10. Okay, so virus hunting is a really bad idea for a couple of reasons. Most of these viruses are never going to intersect with urban human society, no matter what. So you're clawing all of these things that are in great, magnificent isolation. If you get enough of them, at some point, one of them is going to be really wicked and awful. Okay? They were going to get 15 ,000 viruses, a program, a virus hunting program that preceded COVID and fed into the Wuhan Institute of Virology and that very many smart people think could have been implicated in the rise of COVID.

30:01Rob Reid:But many smart people dispute that. I don't want to get into the Wuhan debate. I have thoughts that I've given a great deal of consideration. But leave that aside. If you're going to go in and set a record by doing 8x that, get 15 ,000 of these viruses, and you're going to go to a lot of unstable developing countries to do this throughout Africa, throughout Asia, throughout Latin America, and also part of the program, because this was like a time when it was considered to be imperative to give people of all political persuasions all the tools that science could create. Part of it was to train the hell out of the local scientists to continue to virus hunt and to continue to do some of the other things we're about to talk about after their virus hunting because there was this peculiar idea that it was an unambiguous good thing to inhale all the viruses that we could from these isolated places and put them in these leaky vessels in dense urban centers called laboratories.

30:58Rob Reid:And I say leaky because every category of laboratory all the way up to BSL-4, biosecurity level 4, which is the highest, they all leak. They all leak, yeah. Because they're staffed by humans and we are fallible. And the list of ghastly pathogens that have leaked out of even BSL-4 labs, which are the highest level of security, is dizzying and frightening. And BSL-3, don't even talk about BSL-2. That's a couple steps above a dentist's office, right? Okay, okay. So anyway, it was going to inhale all of these bugs. And then it was going to carry out a series of experiments called a characterization, which to strip down a lot of complexity basically means they were going to figure out which of these had the highest odds of being true weapons of mass destruction.

31:41Rob Reid:Things like COVID killed at the scale of a world war. Things like COVID plus, 15 ,000 shots on goal. You're probably going to find quite a few things. Let's say you find seven or eight of these things. Nobody can predict exactly how many they were going to find, but there was pretty widespread, I don't know whether to call it optimism or pessimism or confidence, but widespread belief. There would be at least a handful. Step three, craziest one of all. You take the genomes of those things. And remember, there's a lot of smart people who can assemble a live virus just from the genome. And you publish them to the entire world.

32:15So we're digging up viruses that are heretofore undiscovered. Yes. and dragging them into a lab in an urban center for analysis and then saying, here's, by the way,

32:26Rob Reid:what are the worst ones? Oh yeah. We're filtering for the worst ones. Let's really find the worst ones. And then publishing what they look like from a gene structure online for anybody to just take a look at. Here's the recipe. And when you say here's the recipe in 2023 or 2024, particularly with a coronavirus or an influenza virus, at that point, you're telling 30 ,000 unvetted strangers across dozens of countries. I have a feeling North Korea is one of them. Do what you will. And then again, let's go back to the passage of time. You're not just telling these 30 ,000 people of whom most probably are actually authentically stable geniuses.

33:01Rob Reid:Yes. Not all, but most probably are. You're telling 300 ,000 people seven and a half years from now, because seven and a half years from now, or maybe it's 17 years from now, the number of people who can do that stuff is going to grow by an order of magnitude because the tools are going to get It's simpler and better and cheaper, just like computers. And you're telling 3 million people in 15 years. This is the height of lunacy. We really regard it as a suicide pact for humanity. What happens? Let's say you find seven of these things. And let's say somebody out there is really bent on destroying the world.

33:37Rob Reid:There are people like this. Look at the Vegas shooter. Killed 50 or 60 people. He wasn't going to stop. He stopped because he was shot. I mean, there is a certain number of people, more than one a day worldwide, who decide that they're going to die in the act of killing as many people as possible. More scary to me is a group called Aum Shinri-Q. Is that the Japanese cult that gassed a bunch of people on the subway with sarin? Sarin. Sarin gas. Yes. So this is in the 90s. It was a doomsday cult in Japan. It can be easy to dismiss them because I think they killed a handful of people with a sarin gas attack.

34:11Rob Reid:But they were an organization with about a billion dollars in assets. Really? Tens of thousands of adherents. They were even more popular in Russia than they were in Japan. I just saw a documentary a couple of years ago. I had no idea because I remember seeing a photo of the leader and going, okay, people followed this guy. Were there a dozen of them? No. Oh, my gosh. Many, many, many, many thousands. And when they raided their – they had people working in bioweapons. They had people working in chemical weapons. They had some PhDs that were part of this. This is insane. When they busted their compound, they found a Russian attack helicopter.

34:47Rob Reid:How do you get that to Japan, right? What? That's insane. Apocalyptic ideology. We're going to cause the, you know, like lots of people, a lot of religions are really eager for Armageddon to come along. Man, we got to start a cult. These guys are - I know. This was quite a profitable business. I want a Russian attack helicopter. I know. Sounds fun, doesn't it? It doesn't sound like a blast. I don't want to use it. I want to use it. I want to use it for self-defense. Strictly self-defense and who can deny me, I'm pretty sure. Anyway, so you look at an organization that's sophistication existing in today's environment and the ease with which that they could obtain tools and techniques, just so far, far beyond.

35:27Rob Reid:So you have to worry about these things. So let's say somebody like that decides, I'm really going to mess things up. Deep Vision came along. Thank God for science. I have, you know, seven weapons of mass destruction that I can switch on at will. COVID came to us. I calculated this and I forget the exact. God, I wish I remembered the statistic. But it came to us. It emerged in Wuhan. We had very early warning. It basically, by the time it hit the West Coast of the United States, it came at a rate of about four and a half miles per hour. That was the amount of time for that virus, even with all the international travel and so forth.

36:02Rob Reid:Imagine if somebody is seeding this in seven of these things in 20 international airports all at once. Right. So it's the slow motion explosion, but from multiple points of origin. It is coming in from every direction. And it's not just one thing. It's seven things. And remember how long it took to get a COVID diagnostic. That is insane. That is the absolute worst case scenario. I do like to think and authentically think it probably wouldn't have happened. And lone wolves scare me more because there are far more of them and they won't be super empowered this year. But, you know, with the passage of more time, they will be.

36:38Rob Reid:And then the last and probably the least expected one is what I'll just call chaos agents. You know, kind of like the people who hijack your hospital and say they're going to turn off all of the ventilators, but don't really want to. They just want their two million bucks. And they can go to sleep at night because, like, I'm not really going to kill those people. I mean, I will if they make me. So an example I have to think about is this. Imagine all those genomes were published. Everybody's sitting here saying, oh, my God, I hope nothing bad happens. 90 days go by. It's kind of quiet. And then all of a sudden, Iran, let's say, says, you know, hello, people of the world.

37:14Rob Reid:This is kind of incredible. We're really grateful to these scientists. We've booted up a bunch of these viruses, and we've been testing them in the dungeons where we keep our tens of thousands of political prisoners. And it's pretty amazing. They have a case fatality rate of 30%, very, very contagious. And because we love the world and we love peace, we have created a bunch of petri dishes of these things. And we put them in our embassies throughout the world with a double helping at the UN mission in New York City. And we never intend to release these. But boy, if something goes wrong, that is a deterrent that's much more powerful than a nuclear bomb.

37:49Rob Reid:And they could be lying. If they said that tomorrow, it would be on page 36 because nobody would believe them. But if we had this advanced scientific establishment saying, this is how you do this thing, you couldn't disbelieve anybody from that day forward. So we decided this program had to be shut down. It's crazy to me that anybody thought it was a good idea at first, but I guess it probably didn't look at the other side of the coin. So how did you get this thing shut down? I mean, I assume it wasn't just you banging on doors in Washington. No, it was me putting out a podcast episode and hoping for the best.

38:23Rob Reid:So I got wind of this. I'm a sci-fi writer, as you know. And one of my novels had a section in it about synbio risk, synthetic biology risk. And I got really concerned about that. And I started the podcast as sort of a spinoff of the novel. And I did mostly scientific interviews. And I explored this more and more. And through various people that I met and so forth, I got wind of this program, which wasn't classified. It was part of USAID and really freaked out when I read their sort of basic documents and was very rapidly introduced to this biologist at MIT, Kevin Esfeld, who had gone very deep into it.

38:57Rob Reid:And so he and I – I mean, he was vehement that this thing needs to be shut down. How do we do it? We're not government people or anything like that. And so we did an interview, which our friend Sam Harris was – I brought him on everything that was going on. And, you know, his audience is 20 times the size of mine at the time. Yeah, sure. And he said, like, look, interview Kevin and I will put it out to my own audience. So it got out to a lot of people. And we basically, if people want to listen to it, it is a very, very methodical, fact-based, very accessible. We're careful to explain things in ways that, you know, are a non-expert.

39:32Rob Reid:I'm a non-expert. If I can understand it, I can explain it to somebody else. It's sometimes bashing my head against the wall for a long time to get to my own non-expert understanding. But once I do, so Kevin and I think I did a great job of that. Sam puts it out there and fortunately does it about a day before, like literally, the Ukraine invasion. And that was particularly problematic because we were really hoping to influence a woman named Samantha Power, who at the time was running USAID, of which this was a tiny little program. And we had ways of getting to her. But once the Ukraine invasion happened, I'm sure she was 36 hours a day dealing with Ukraine relief efforts.

40:13Rob Reid:Quite likely. So it was crickets for a little while. And then a couple people reached out to me who had heard the episode. And we kind of put together a little scooby gang of people who are going to try to work on this. The main folks, I'd say, and I hope they don't find this embarrassing, I do cite them, and they try to keep it on the DLs, a group called Helena. They're an interesting thing. They are kind of a white paper writing, influence generating. They don't like to be called lobbyists, but they do their best to influence public policy. They're very effective at that. They know a lot of people in government.

40:47Rob Reid:But that's half of what they do. The other half is they actually run a venture fund and they invest in world positive companies. They reached out to me and we ended up teaming up on this. And I'd say they primarily ran with the ball from there. The most diverse people we were able to bring into this. We got significant help from Chelsea Clinton and Lindsey Graham. Okay, this is quite the—

41:37Rob Reid:completely defanged quietly. And then kind of to our astonishment, about a year later, it was officially and pretty summarily killed. And I think that sent out a big signal in public policy and public, quote unquote, health circles that certain types of crazy on the edge science are, I can't say is not going to be tolerated anymore. It's not going to be funded anymore. We're nowhere near as far along as we need to be. But it was a pretty big sign that things like virus hunting, things like virus characterization, things like gain of function, which was not part of deep vision, but it's adjacent to that kind of really let's risk it all.

42:15Rob Reid:So we might be able to get on the cover of nature or science type of mentality really was dampened for quite some period of time after that. So gain of function is where you say, hey, this is a really deadly virus. Let's make it more contagious just to see if we can. Make it more deadly, make it more contagious, make it, you know, make the incubation period longer. You ask why would people come up with ideas as awful? And the answer is there's a thicket of beliefs that surround this. And I think the people on the other side generally have the best of motivations, which is one of the scariest things about this, because you think about if they can come up with an idea this bad, imagine what a bad person can, right?

42:54Right. Because surely there's somebody who was punching the air and cursing your name when their program got shut down. You don't understand why this is a good idea. But I would love to hear what you think that person was thinking.

43:05Rob Reid:Sure. I mean, one of the most appealing on the surface, but kind of logically bankrupt ideas it's put out is like, we can make a vaccine. Okay. So what you're going to do is you're going to pull this virus out of a cave that probably never would have exited that cave. And you're going to say, well, long shot chance that it did, why are we going to have this thing? Will yada, yada, yada make a vaccine? Here's a problem with that. Yeah, this thing is really dangerous. Boy, it would be really bad if it got out. Wow, we've made it very famous by publishing the genome. It's now in 170 labs throughout the world because everybody wants to poke and prod at it.

43:39Rob Reid:A little more likely it's going to get out. Oh, but you're working on a vaccine. Here's what happens when you're working on a vaccine. You get a promising signal. Now you have to test it in humans. How do you do that if the pandemic hasn't broken out? Are you going to infect 100 people with this pandemic virus or pre-pandemic virus that might have a 30 % chance of killing them and give half of them a vaccine, which may or may not save their lives. No, you're not going to do that. That's called a challenge trial. We couldn't even bring ourselves to do that during COVID when people were actually getting naturally infected.

44:08Rob Reid:Long story. But no, that's not going to happen. So you're going to know nothing about safety or efficacy. Okay. So now, thanks to your efforts, there is this incredibly fulminant pandemic. Now we can start testing people. Hey, great, it worked. Oh, go us. We've figured this thing out. Now you have to ramp production. Remember how long that took during COVID? Exactly. Most of the ramp time was also regulatory, at least safety and everything else. The fantasy is you discover this thing and you have 8 billion vaccines ready to go. Yeah, that doesn't make sense. The pandemic busts out. But no, you're not going to have that because you haven't tested it.

44:43Rob Reid:What are you going to invest? $70 billion in a vaccine? It's insane. So there'd be arguments like that. There'd be arguments around a very well-intentioned and usually very beneficial idea, which is called open science. And open science, this is a grotesque abbreviation, it basically says everybody in the world has equal right to all the knowledge that is gained by science. And many, many positive things come from this. But if you become an extremist about that, you know, like people said during 9-11, the First Amendment is not a suicide pact, right? So part of it was like, well, if we do find these annihilating pathogens, who are we to say Kim Jong-un shouldn't have it?

45:25Rob Reid:Right. Who are we to say that the next Osama bin Laden shouldn't have it? Isn't that a little bit elitist? Isn't that a little bit Western centric? Right. Right. Who are we to say that the next gen of Columbine kids? Isn't that a little bit ageist and obviously exaggerating? But that is the kind of thing that would create this incredible pressure to once we've discovered something that could have life saving benefits. Everybody needs to have it. I mean, I suppose if you find a vaccine for something that already exists, yeah, maybe you should share that because we need to make sure the whole world doesn't get it.

45:58But sort of increasing the odds that the world gets it by a thousand percent or a million percent. Orders of magnitude. Right. In order to maybe have a shot at fixing the problem that you basically have created, it does that logic completely falls apart.

46:11Rob Reid:And if you were marinating in a stew of folks who think like this, and then there's this other thing, which is sort of like the Ramones side, the bad boys side, like the people who work in these areas are kind of like in some eras and in some institutions have kind of been the cool kids. Yeah. Like, oh, wow, he's making this virus that could kill everybody. And when you talk to him, he has these really articulate and slightly pompous, but kind of sexy arguments about how only a knucklehead wouldn't think this was great. Right, right, right. So if you're marinating in this area where gain of function is good and only the dummies don't understand it and, oh, yeah, even if this is another big thing, even if the bad guys got it, we're so much smarter than them.

46:53Rob Reid:Right. We would, you know, we would come up with an antidote, boom, boom. When you're marinating in that stew and then you have this sort of, you know, very politically correct view of open science, et cetera, et cetera, you can talk yourselves into some pretty crazy things. If you're thinking, surely someone responsible is carefully monitoring all of this, congratulations on maintaining a childlike sense of wonder. We'll be right back. This episode is sponsored in part by SimpliSafe. Home security isn't really about the house, it's about the people inside it. You can replace things, but getting back that feeling of safety is a lot harder, and our family knows that firsthand.

47:27A couple years ago, we had an actual break-in at our home. I cannot overstate how violated we all felt. And to this day, every night before our kids close their eyes, they ask us, is the alarm on? Years later, it's still on their minds. That's why keeping a threat outside before someone even gets close to the door matters so much to me. SimpliSafe doesn't just protect your home. It protects the life inside. Unlike traditional systems that only alert you once a break-in is underway, SimpliSafe is built to stop it before it starts. Its AI-powered cameras watch for potential threats and alert a live U.S.-based security specialist.

47:56Their agent intervenes in under 30 seconds, speaking through the camera to stop an intruder in their tracks and calling for urgent 911 dispatch when needed. That ability to intervene before someone gets inside is what stands out. Join the more than 6 million people who entrust SimpliSafe with their home security. Right now, save 60 % off with professional monitoring when you visit SimpliSafe.com slash Jordan. That's SimpliSafe.com slash Jordan.

48:20Rob Reid:And when you order, please write in the Jordan Harmager show in the checkout survey. It lets them know we sent you and really helps support the show. SimpliSafe, ahead of crime in real time. This episode is brought to you by Xero. That's Xero with an X. Running a small business means there's always something competing for your attention. Managing your finances shouldn't make your day more complicated. Xero brings accounting, payments, payroll, and analytics together in one connected platform. So you spend less time piecing together the numbers and more time actually building your business. We've been using Xero for nearly a decade now, and honestly, we don't know what we'd do without it at this point.

48:53That's meant clear reports, dashboards we can actually read, forecasts, and AI-powered insights that help us understand our own cash flow and see how the business is really doing. It was simple and easy to use from day one, and it still is. In fact, 91 % of customers surveyed agree that Xero is easy to use, and when you choose Xero, you're joining 5 million customers worldwide, including us. That's Xero with an X at Xero.com. The 91 % claim is based on a 2024 24-0 survey conducted in the US, UK, and Australia. If you like this episode of the show, I invite you to do what other smart and considerate listeners do.

49:27Take a moment, support the sponsors. They make the show possible. All of the deals, discount codes, and ways to support the podcast are searchable and clickable at jordanharbinger.com slash deals. If you can't remember the name of a code or a sponsor or whatever is not working for you, email us jordan at jordanharbinger.com. Someone will surface that code for you. It is that important that you support those who support the show. Now, back to Rob Reed. We really are not in a technological place where we can stop a pandemic cold in its tracks. It would be different if the whole world had some level of like early virus detection because of something that was on your phone.

50:03Right. And then most people living in a middle class Western household could maybe print a vaccine and we could get a billion doses. We could get to Africa within a week.

50:13Rob Reid:Every doctor's and pharmacy can print these at scale. Yes, exactly. Because we cracked the code in 16 hours because this biosurveillance program picked it up in Chile when it leaked out of that lad. And, you know, we're watching the Super Bowl three days later. We've forgotten about the whole thing. Right. The biggest problem is that CVS made 100 ,000 extra doses and now they have to write off a loss in their shareholders or something. And by the way, we can get to that sci-fi situation with some years, but not insane budgets. Give me the budget for the train to nowhere in California, that$24 billion or whatever it is.

50:45Rob Reid:like we could establish so much in terms of virus surveillance and robustness. But, you know, they're not crazy budgets. We could live in that world. Yeah, that's interesting. I guess my point is, since we don't live in it, we can't be really rolling the dice on this kind of thing. It's just it doesn't make any sense to do it that way. And it's like these people could maybe take a crazy chance when we have all these safeguards in place, all these systems, all these mitigation solutions in place. But right now it's, oh, yeah, look how we handled COVID, eh, not so great, pretty poorly. Let's just roll the dice over and over, except with way more deadly pathogens.

51:21What are you thinking? Like, this is just not the time nor the place to be messing with this stuff.

51:25Rob Reid:And the risk vector today, just to get a little bit more upbeat, is less that kind of crazy science being practiced in ivory towers, although that's still something you'd be very worried about. It's the fact that the intersection of AI and synthetic biology is really going to accelerate this proliferation that some of us have been worried about for several years and we didn't really see today. Nobody saw today's AI coming. But that really democratizes things and it really proliferates things. And when you have to worry about nine people not hitting the blinking red button, which is the situation with nuclear weapons, plenty scary, but nine is a tractable number.

52:01Rob Reid:When that number starts ballooning, you know, way past 30 ,000, way past 100 ,000, you can't realistically hope that nobody's going to ever unleash something bad. What we need to do while we still have time is invest in those countermeasures, invest in building a society that is a multi-level immune system to counteract this because the odds of something getting out there are around 100%. So when you say AI changes biological risk, I use AI, it answers a lot of questions. What can it do? Can it generate new designs or just help people complete physical work somehow? So I guess what has actually changed in the last three to five years with AI?

52:40Rob Reid:I'd say really in the last, you know, nine months, because the models have just gotten so much better, so much better. I mean, the first thing to understand is life like a novel. There's a language with an alphabet. There's 20 amino acids that, you know, basically make everything in life, everything in the biosphere, not just us. And there are just, you know, four nucleotides that code for those 20 amino acids. So it's like, our English language is 26 letters, it's 20 letters. As in English, there are repeated themes. There are things that work well and work repeatedly well, and there's things that work poorly and work repeatedly poorly.

53:17Rob Reid:And so just very, very big picture. The ability of a transformer model, which is what underlies all of the chat bots and everything that we have, to understand, fundamentally understand a language and start excelling at it in ways that we wouldn't expect a computer to excel at that language is extraordinary. And it's accelerating in a compounding way, much as the language-based AIs that we're all experiencing. And so I'd say there's three things that you'd be concerned about. And it turns out one is, there's evidence that one is not a concern at all, but the other two are. And so thing number one is, you know, the person who wouldn't possibly be able to come up with an assembly program protocol for a virus, let's say.

54:03Rob Reid:They're not going to be a knucklehead. Like somebody who's pretty damn smart, maybe getting straight A's in undergraduate biology, is working in a researcher's lab where there are all kinds of tools and very, very little. These are not micromanaged places. These are places where brilliant people are urged to go run rampant with their intellect and come up with their own ideas. And they're definitely working for somebody, particularly when they're young. But everybody loves giving people autonomy. And these are just not highly managed environments. It's not like working at Google, right? And so let's say you're getting straight A's as an undergrad in biology and you have a very, very dark side.

54:43Rob Reid:Or you don't have a dark side at all, but you just have some intellectual arrogance and you're drawn to the cool, wild, scary thing. I remember when I had chemistry set as a little kid. Do they still give kids chemistry sets? It seems like a bad idea. I don't think so. I haven't seen one. I mean, there's little things that you can do, but it's like flour and vinegar or whatever it is. Yeah, no. These things were a little bit more real than that. I always wanted to do the things. Like, they did have little bottles that had skull and crossbones on them, and they really had them for a good reason.

55:11Rob Reid:And you've always wanted to do the dangerous stuff. Have you seen those old toys where it's like they give you nuclear material and it's in a kid's set and it's like in a glass ampoule? Oh, God. I'm getting on eBay as soon as I get home. Oh, yeah. These are real. And they're from the 50s and the 60s. And it's like, it contains real, I can't remember what it is. Isotomes. Yeah. And it was like - Real polonium. You can look at it and it glows. And like, if you do certain things, and it was just like real nuclear material. We had a radon alarm clock when I was growing up. Yeah, yeah. And it was a really cool orange glow at night.

55:41Rob Reid:We kept it in the guest room. I guess my parents didn't really like their friends very much. Yeah, yeah. Anyway. Is it harmful if it's sealed? Or I don't even know what that is. It's not like it was sealed in lead. Right. I don't know. I've studiously avoided trying to look into it. They sure don't make them anymore. Yeah, yeah, yeah.

56:24Rob Reid:really accelerate their theoretical understanding and their understanding of protocols, assembly protocols and other things, and get their own creativity going to work. And like one of the most exciting things for me to do with an AI is to get my creative mind running down a path that I, you know, hopefully no other Claude user or very few have gone down and have that amplified by Claude as a research tool. It's like the combination of person and machine, like a centaur, is one of the most exhilarating things about AI. And I'd imagine that biology is your jam. So it's two things. It's rapidly up-leveling the knowledge that somebody can gain about things that could be dangerous without having to go to a professor about it.

57:06Rob Reid:Now, they're going to run into what you ran into, which is called a classifier model. Classifier model is like an independent model that stands between you and the frontier model. It's like a bouncer in a club. But as you found it, there's workarounds. Every system card that any frontier lab has put out, when they release a new model, they have something called a system card, which speaks very candidly about the model's capabilities and limitations and so forth, every system card that's ever been published has acknowledged that classifier models are very, very inadequate. And there's all kinds, I mean, they do their best and they're getting better, but they can't preclude any indetermined person from coming up with ways to getting around.

57:44Rob Reid:And so that's problem number one. And then problem number two is what you said, is coming up with novel ideas that haven't been thought of before. And that's something that some Some of the more senior people in this field really seem to worry about a great deal. Really significant breakthroughs in terms of lowering the threshold necessary to do something potentially very risky. Yeah. Tell me about cloud labs. These were kind of like, oh, well. If you can't do it, someone else will do it for you. I mean, that was even more scary. Yeah. So this actually gets into the third thing that I said, you don't have to worry about that much, but it mitigates that.

58:19Rob Reid:So what the LLMs and people have done some experiments on this, actually, like very deliberate experiments. LLMs are not particularly good at upleveling somebody's physical skills in the lab. There's actually a lot of dexterity and what they call tacit knowledge about like, how do you do pipetting? How many times do you run this cyclotron at what RPMs and so forth? There's a lot of things that you just get to know by being in this very complicated environment with, you know, 17-ish core skills that all take a couple years to learn. But at some point, you're fluent in all of them. You're putting them together.

58:54Rob Reid:That would really elude, you know, like the 9-11 hijackers would have a very freaking hard time doing anything in a wet lab. Right, yeah. They would not have been able to pull something like this off. But these cloud labs are a really interesting thing. And I think they have unbelievable potential. What a cloud lab is, is it's basically an entirely robotic lab, almost entirely or truly entirely. And laboratory automation is going through a great renaissance right now. And so you still want to have a lot of graduate students working for free, pipetting in your lab if you've got a research grant and you get to do that sort of thing, and they're part of your academic environment.

59:31Rob Reid:But maybe you don't have that sort of environment. You were probably not going to be able to get access to a great deal of equipment and know-how five years ago. But now we're in the dawning days of cloud labs. And what they can bring together is many tens of millions of dollars worth of equipment under one roof that anybody gets to access. And that is something that even a very high-powered academic scientist with great deal of tenure and incredible grants and so forth could very easily be green with envy at the amount of equipment and the quality of it inside of a cloud lab. Why are they able to do it?

1:00:07Rob Reid:Because they have many, many customers, many of them thousands of miles remote, saying, I want this particular assay run. I need this very expensive mass spectrometer that no one lab could afford. it. But like, guess what? The cloud lab has got it busy 20 hours a week because everybody wants that mask back. And now it's there, right? And a lot of other equipment. And so, you know, we can't even figure out how to prevent deadly strands of DNA being generated and sent in the mail. We're doing our best and then the screening protocols are getting better. And thank God, there are people calling it in Congress for it to be made mandatory, but it's going with bizarrely glacial speed.

1:00:47Rob Reid:But that is such a simple thing that we are unable to control. Now imagine you're in this, and remember what my friend Kevin Esfeld did by cutting up his 1917 flu genome order across dozens of different labs. I'm going to do a little thing here, I'm going to do a little thing there. Nobody's going to put two and two together. Now take that from not merely assembling a long strand of DNA to doing extraordinarily extraordinarily involved experiments where you can have this little piece done here and that little piece done there. I don't know how you audit that. I don't know how you protect against that.

1:01:23Rob Reid:But I will bet on the innovation and the intelligence and the self-interest of the American scientific and entrepreneurial community to crack that problem. But it needs to be a priority from the very, very beginning. I think that humankind could reap extravagant benefits from cloud labs. But if we take no protective measures as they start to spread, it's going to be a pretty lousy trade. Yeah, it's great that people can access tools that they otherwise would not have. I mean, you could have a bioprogram at a small, whatever, scientific school that just has underfunded. Sure. Maybe it's in another country, and now they suddenly have this gift.

1:02:02Rob Reid:Just doesn't have equipment. Has great theory, doesn't have equipment, now they do. Yeah, that's great. But yeah, of course, if they're just also letting, I don't know, teenagers in North Korea use this stuff, or teenage hackers in North Korea, bad actors use it, and their verification process is email us a photo of you, that's going to be a tough thing. I mean, I kind of think of it in a weird way, you know, in terms of like hijacking and aviation. I mean, we could have had, the world could have had global aviation without 9-11. Yeah. 9-11 was optional. 9-11 was an abject failure of stupefying proportions.

1:02:37Rob Reid:You know, the budget to harden cockpits, which was the one thing that would have made all of that stuff impossible. So, you know, that was not a tradeoff. We could have had global aviation and all of its benefits without that. But it was crippling miserliness, a catastrophic failure of imagination, you know, preposterous overconfidence that now we could pull off something like that, et cetera, et cetera. And it turned out that the fixes post 9-11 were, in the scheme of things, so goddamn simple and so unbelievably low budget compared to the trillions that got burnt up in the fires of multiple wars and stupefying budgets and blah, blah, blah, blah, blah.

1:03:19Rob Reid:It really is one of the worst lapses of vigilance, intelligence, just baseline competence that the world's ever been through. And we can have the benefits of synthetic biology, cloud labs, and so much more. We really can. And those are benefits. I don't want to live in a world without cloud labs. I'm real excited about them. There's one called Emerald. It used to be in California. It's in Texas. And a friend of mine knows the founder and CEO. And I'm hoping to meet him sometime this fall. But man, we need to treat them with care. The way that we can screen for pathogens is actually quite interesting as well.

1:03:54It's just the wastewater treatment. I mean, they already do some of this, right? They do some. Wastewater. Yeah. And I mentioned something on your phone. I don't know what that would be, like an air sampler. I mean, there's a future in which all of this is pretty standard and not that complicated. There is.

1:04:06Rob Reid:And wastewater alone, I mean, just picking up the wastewater in airports throughout the world, really low tech, kind of unpleasant. But you can learn so much from that, or we could learn so much of that. But if we put tens of millions of dollars a year into it, the countermeasures are really, really affordable and well understood. How would these DNA places screen? I mean, because you mentioned you can just order two times instead of ordering. I need smallpox and I need this and I need it. And they're like, whoa. But then you just place two orders with two different credit cards. It's pretty achievable.

1:04:37Rob Reid:And so there's a program called Secure DNA, which has been proposed. OK. It would be free to all users. It's hard to argue with that. And it's kind of like a much more sophisticated version of this thing that we were talking about earlier. And so its use relies on the notion that it's ubiquitous. If it's free and if it's mandatory, you can get ubiquitous pretty quickly. The two entities in the entire world who have the most to lose and therefore have the most to gain by making something this inexpensive and commonsensical universal are the United States government and the Chinese government. Right.

1:05:12Rob Reid:Sure. Right. We have more to lose than anybody else. Why don't we get together? We need an easy win, something we can all feel good about. Can't think of anything better than this. So what does secure DNA do? Good point, because I think China realizes that COVID really messed them up and they didn't handle it well. I think they probably have to admit that by now. Not great for the global reputation. No. Lots of other downsides. People getting welded into their apartments and dying of starvation was not a good look. Trillions in economic damage to their economy and ours as well. You know, it's like this should be a pretty easy thing to agree on.

1:05:40Rob Reid:And the way that the secure DNA protocol works, and I'll simplify it in some ways and also I don't fully understand it. But the point is you have, first of all, little, little segments of DNA that can be assembled into really major ones, not allowed. And the thing is, like, you know, there are so many thousands of base pairs in a virus. And four, you know, you multiply. If something squares all the time, you multiply by a factor of four because there's four nucleotides, right? You have four, 16, 64, boom, boom. When you get to a strand of 25, you've exited the realm of random possibility. Right, I see.

1:06:16Rob Reid:This one looks a little bit suspicious. Now, it's replicated in a bunch of other things, but what else is going on on the network? Do we seem to have a self-assembly going on? Pretty easy to determine because the whole thing is encrypted. And nobody really knows. The system knows. But you don't have to sacrifice if you're a client. Your stuff is coming in encrypted. It's being judged and it's getting bounced if it needs to be bounced. But nobody on the other side is seeing it. Nobody's stealing your IP. Nobody's watching over your shoulder. The other thing that's really important about it is that people can submit.

1:06:52Rob Reid:New dangerous strands of code are going to become evident at all times. And what you want is something that rapidly implements them. And so basically has a mechanism of submission and something that's a little bit like peer review to make sure that people aren't just spamming the system to get, okay, this is, you know, as of Tuesday afternoon, this is a really dangerous sequence. And it is remarkably easy to do this. And it would cost literally nothing because this is a not-for-profit. And they're providing their tools and their software and their know-how to the world for free. Yeah, wow. I mean, look, a smoke alarm is useful because I know what to do when it goes off.

1:07:28So what's the equivalent action here if we detect a bunch of pathogens in airports?

1:07:33Rob Reid:Yeah, well, I'll actually roll back a little bit and just talk about, I'm going to go to hijacking again. In the late 60s, early 70s, this country literally had 40 to 50 major commercial jet hijackings per year. Yeah, so like one a week almost. Like one a week almost. It was insane. And part of the reason that we had this was the airlines were fighting tooth and nail against airport security because they didn't want to pay for it. And they also said we think that it might make passengers think that flying is dangerous. Well, I'll tell you what makes passengers think flying is dangerous. Landing in Cuba instead of that.

1:08:08Rob Reid:It is a weekly hijacking. And you got Cuba right. These were kind of hippie hijackers. And so a very large majority of them demanded to go to Cuba because they thought Castro was going to welcome them and give them a cigar or some God knows what. Gave them a machete and told them to harvest sugar. Yeah, if they were naughty. And so it got so bad they literally, literally, and I think I said this on Sam's show, had a dormitory for wayward American hijackers in Havana. And they tried to make it as unpleasant as possible. But then what happens? Okay, finally, there was a hijacking in late 72, and the hijackers were talking about crashing the airplane into a nuclear power plant.

1:08:45Rob Reid:Oh, wow. And this got out in the press, and it got security, and intelligence was aware of it. And finally, the government said, shut up, airlines. They implemented very, very baseline hijacking or security at the airport. Metal detectors or whatever. Yeah, basic things. 1972, 50 hijackings, 1973, zero. Wow. And that it stayed at that unbelievably low level until 9-11 happened. And we realized we'd taken our eye off the ball for decades. And then we dialed things up again and we have not had one in 25 years. That's a good point. So what does this tell us? It doesn't tell us that hijacking is impossible.

1:09:19Rob Reid:You can't make it impossible. Right. What it has done is it's raised the bar for doing it to such a high level that only relatively large-ish and very functional and very ambitious groups could even hope to pull it off. And when you suddenly have dozens, even hundreds of people collaborating on something like this, that's when intelligence can bust it. That's when law enforcement can bust it. And so something like secure DNA, it really raises the bar. And I think the person who might have had a bad night and is really smart about a few things in bio and decides, I'm going to knock out Cleveland tomorrow, he just can't.

1:09:55Rob Reid:You know, the impulse goes away. The kind of rookie who's kind of smart goes away. You just you raise the bar a great deal. OK, so now let's say something is happening and we don't know what it is. And the worst possible thing we can think about is what's called a stealth pandemic. A stealth pandemic would be one in which somebody has engineered the incubation period. This was actually in Jacobson's book. Had the longest incubation period, I think was part of what she had in that bubonic plague thing. Good news. There are realistic ways to make engineered pandemics dramatically harder. Bad news.

1:10:28They require governments and scientists around the world to coordinate competently. So enjoy these commercials.

1:10:37You're listening to this podcast, so I know you've got a curious mind. Here's a helpful fact you might not know yet. Drivers who switch and save with Progressive save over$900 on average. Pop over to Progressive.com, answer some questions, and you'll get a quick quote with discounts that are easy to come by. In fact, 99 % of their auto customers earn at least one discount. Visit Progressive.com and see if you can enjoy a little cash back. Progressive Casualty Insurance Company and Affiliates. National average 12-month savings of$946 by new customers surveyed who saved with Progressive between June 2024 and May 2025.

1:11:13Potential savings will vary. Ever notice how life's best stories don't happen in your living room?

1:11:20Rob Reid:They happen on the open road, out on the water, or parked under the stars. At Progressive, they get that you want to focus on the experience, not worry about the what-ifs. That's why they offer quality insurance designed for your ride, whether that's a boat, RV, or motorcycle. Adventure with confidence. Visit Progressive.com and see how easy it is to protect your favorite way to get away. Progressive Casualty Insurance Company and Affiliates. Not available in D.C. Prices vary based on how you buy. Come join us on the Jordan Harbinger subreddit. A lot of discussions there about all of these episodes.

1:11:53Me, Gabriel, Bob, and the whole team are in there as well. That's over there on the Jordan Harbinger subreddit. Now, back to Rob Reed. An incubation period is the amount of time it takes for the virus to show up in a human symptoms wise. And you want that to be long if you want your virus to be contagious because someone spends two weeks infecting a bunch of other people before they know what's going on.

1:12:13Rob Reid:That is the worst case scenario. So let's imagine that somebody is inflicting that kind of a pandemic on the world. And we have spent a very low number of billions of dollars on creating an unbelievably robust global surveillance system that is definitely sequencing all of the wastewater in all the major international airports. And it's also doing some aerosol sensing and a whole lot of other things. You're going to get signal very early because here's what's going to happen. When they do deep sequencing of wastewater, they find an enormous—it's a very, very tiny sliver of the nucleic acids in there.

1:12:47Rob Reid:How do they find a strand of DNA in a million people's poop? That's impressive. You don't want to know. Yeah. You really don't want to know. But they wear gloves. Let me just leave it at that. Yeah, they better. Definitely wear gloves. And so what they would find is like, here are some DNA shards that we've never seen before. That's interesting. Holy cow. They are 10 times more widespread than they were yesterday. Yeah. They are a hundred times more widespread than they were the day before yesterday. This can literally only mean one thing. It means something is contagious and it's growing very exponentially.

1:13:26Rob Reid:And you are aware of this when dozens to hundreds of people, maybe hundreds of thousands of people are infected. And if you put your surveillance system together in an intelligent way, and I'm certain we would, you'd probably be able to pinpoint geography very, very quickly. I see. This is coming out of Uruguay, you know, because look at the pattern, you know, from, you know, to Buenos Aires to Miami, blah, blah, blah. And with that kind of warning, you could get PPE out to all of your frontline workers. So breathers and masks. Yeah, like very higher quality PPE than we've had before. And there's a really interesting not-for-profit.

1:13:59Rob Reid:It's working very hard on this with excellent funding to come up with the idea of like, who are all our frontline workers? They number in the high single digit millions. The people you need to keep the lights on. You know, it's not all cops and fire people. It's, you know, people with power, you know, people delivering the food. You know, imagine a Mad Max scenario in which suddenly people are dying like mad. No frontline worker would dare to go out because they might infect themselves and their whole family and kill them all. The food gets shut off. The water gets shut off. Law enforcement gets shut off.

1:14:27Rob Reid:The power is out within six or seven days. And we've got about nine days of calories on all the grocery stores in the United States at any given moment. Wow. Yeah. Whereas if we have had the foresight to say, yeah, we know who all the frontline workers are. We have 10 million suits that didn't cost all that much, but damn, they're good. People have been trained in putting them on. We've gotten this signal really, really early. Everybody go home. We got 10 million of these things, which are well short of bunny suits, but they can do the job. The food's going to come. The power's going to stay on.

1:15:01Rob Reid:Take up backgammon. We're going to get through this together. And that is, we're talking a sliver of the cost of, we may be talking about two days of the defense budget. Yeah. Put that together. And again, if you have international cooperation, because nobody has more to lose than we in China and we just get together on this like, yeah, this would suck. This is why you can't take people directly from unawareness to despair or denial to despair because in neither none of those states, they do anything. And there is actually a very strongly hopeful message here. We talked in 2021, I think it was crazy, about eventually producing vaccines at pharmacies or even at home.

1:15:37And I brought that up earlier in this conversation. Where does that idea stand now? What would that look like?

1:15:42Rob Reid:We let a good crisis go to waste, as they say. With COVID, COVID should have been the wake-up call that unified everybody, brought us together, and realized that like$60 billion a year in the federal budget is nothing, and you wouldn't need anywhere near that. $10 billion. Let's get our act together. As far as I'm aware, nothing, no work has been done in distributed vaccine production. And that's the kind of thing that I would like to see DARPA, you know, NIH and others putting real money behind because you really need to start with blue sky research, the kinds of things that are best financed by governments.

1:16:15Rob Reid:And, you know, no government on the planet has invested more in pure science than we have. It's why our academic science is so incredible, even if it sometimes gets moronically politicized and shoots itself in the foot. There's very, very good talent out there. That's something we should be investing in a great deal. Another thing we should be investing in is ultraviolet light. of different kinds. Germicidal ultraviolet light is one of the cheapest and least appreciated weapons that we could have in making ourselves robust against pandemics. And there's two spectrums here. It's very interesting.

1:16:48Rob Reid:So there's UVA and UVB. There's the things that give us sunburns. Then there's UVC. UVC would kill us, but for the fact that the atmosphere completely deflects it. But we can make it's very carcinogenic, very cancer causing. But it also just nukes bugs. Yeah. Viruses. It just destroys the DNA of poorly protected viruses and bacteria. So UVC has been used in public health settings probably since the 50s. You need to shield the light from people. And generally you put it on when nobody's around. Is this what escalators have when they say like this handrail is being disinfected by UV light. You know about this?

1:17:27Rob Reid:Yeah, and I hope they do it when it's under. I assume so because you can't see it. Yeah, yeah. So UV is often present, not very often present, but it can be used in sort of like the pulmonary system of a building, in air ducts in some places. People have experimented with that. It's very frequently used to disinfect operating theaters. So after a surgery is over, they switch on the lights. That's great. They'll irradiate the surfaces. So all the fomites, you know, The surface stuff is gone. Wouldn't it be great if you could flip on UVC when all the passengers are on the bus and they're breathing on each other?

1:18:01Rob Reid:Well, you give them all cancer, you can't do that. So they irradiate the buses, particularly in China at night. But it's like you kind of flush it out, but people are still respirating. And huge, huge disclosure here. I'm a venture capitalist. and one fascinating company that we've backed called a lumisoft there up in canada has a really cool what they call upper room uh guv upper room germicidal ultraviolet light great it's like a fixture you can put into a conference room or a patient's room the light uh is sent upward it gets absorbed by walls and ceilings and so forth so you're in a room and you sort of see this faint purple on on the ceiling but it's not touching you and they have these very powerful fans that circulate it.

1:18:38Rob Reid:So it's like something like 30, 40, 50 air exchanges per hour. In the room. Wow. In the room. And so now you're getting something that could actually really, really dampen transmission. Is it loud? That's crazy. No, it's silent. It's pretty practically silent. Yeah. It's like a little hum, like you hear an air filter in the room and you sort of hear it. Air is exchanging almost every minute. I'm thinking like, that's a loud ass fan. Well, you're not going to want to wear a hat. It's going to blow off. No, I'm kidding. It's nothing like that. Oh, really? Okay. It's remarkably subtle. I've been to their conference rooms where they have it going on, and I never caught a cold.

1:19:07Rob Reid:Never caught a cold. Never caught a cold. Sometimes I go to these meetings and deliberately infect myself with COVID to see if I can get anybody else sick. I've never succeeded. So far, so good. So far, so good. Gosh. Okay. That's cool, though. It's a cool device. It's cool. And I think that we're going to see some really, really major implementations of this that will finally allow some high-end academics to do A-B studies on what's the ROI on this in terms of diminishing hospital-acquired infections? What is the ROI on this in terms of diminishing sick days? That kind of thing. So this is something we can do right now.

1:19:39Rob Reid:And there is a moment, I think, that we're entering right now where a lot of people are going to do some pretty significant work with this. And I believe that the data that comes out is going to be a pretty major signal to public health throughout the world. Now, the really exciting thing is what's called FAR-UVC. Far UVC is a particularly narrow band of the UVC spectrum where the photons are equally deadly to pathogens, but they don't harm us because they get immediately absorbed by any protein, which means the dead skin cells and even the exteriors of our eyes. And so there's a scientist, Dr.

1:20:15Rob Reid:Bremer, at Columbia who really kind of, quote unquote, discovered far UVC around 2017, 2018. And, you know, it's been mainly laboratory, but I believe there have been some human trials now as well that have shown that as far as we can tell, it's completely benign to living creatures, to mammals anyway. Right, right. Large living creatures that might not want to kill. So that is something that's unbelievably promising. The problem is it's very difficult to make those photons for a bunch of reasons. Think of this as being, I don't know, like vacuum tube radios. And we're about to enter like the semiconductor era, or we will over the coming years.

1:20:51Rob Reid:So the current technology is something called an eczema lamp, which some people call a plasma lamp, which is a more descriptive way of putting it. You've got these two crazy gases that are exposed to a great deal of electricity, and they duke it out. and there's sort of some molecular and even atomic interactions that result in a whole bunch of photons, some of which are far UVC, and then you mask that thing. You filter out all of the deadly light. And what comes through is far UVC, but it's kind of a trickle. So it's not— You need a fusion power plant in your backyard to make it happen. Yeah, yeah.

1:21:25Rob Reid:It ain't there yet. But, you know, very few people would have seen the original computers at MIT and thought Tinder. Right, that's true. We will get there with this. And, you know, I'm aware of some near-term developments that might be really, really game-changing in the very near future in this that haven't been announced, but I'm pretty excited about them. And that would be a real unlock because we had far UVCs, and people are talking about one of the designs. I saw you screwing in a light bulb socket. Yeah, wow. Another one, which is second generation, is flat screen. And when you think about that, and we need to do further safety, you know, we need to make sure that we can, you know, handle high doses of this and so forth.

1:22:04Rob Reid:And we're good at that. We're good at phase one, two, three. You imagine the world in, let's say, 2034, and people are aware of this, and these are inexpensive as hell, and they double as maybe even TV screens when you're not in the middle of a pandemic and so forth. And now we get the signal out of Uruguay. Now we've pumped out the PPE to people. Now harmless UVC lights are switching on everywhere. And I get optimistic. Yeah, that's really interesting. I can imagine they need that in provided, safe, kindergarten classrooms? You need it in schools. There's been a lot of conversations about that.

1:22:40Rob Reid:Yeah, certainly kindergarten on up. You definitely need it in healthcare places, long-term care, you know, retirement facilities. Hospitals, it's like the amount of hospital-acquired infections. It's crazy. It is very, very high. And this is the sort of thing that could pretty much eradicate it. Because the problem with upper room UVC, like I said, is you do it when people aren't in the room and then they come into the room and goddammit, they start breathing. You know, like you can't stop people from breathing. And so it's sort of a temporary measure. And if you did something like this company, Illumisoft has, you're still radiating upward.

1:23:12Rob Reid:So you're not taking care of the surface, right? But for UVC, it could be on at all times, irradiating the surfaces, you know, knocking the pathogens as I exhale before they even get to you. So this is the slightly sci-fi near future that we can have if we want it by investing in these technologies, by supporting the companies that are bringing them to market. I guess then the danger is overuse, right? Because you probably should get some colds. Probably, yeah. You probably should get some light disease, but you don't want old people getting a particularly nasty influenza. Yeah, and we'd overcorrect, and then we'd undercorrect, and we'd be bad at it, but then we'd get good at it.

1:23:50Rob Reid:Yeah, yeah. Like peanut allergies. It didn't used to happen until people started saying, no, peanuts until you're five. Right, that's true. Interesting. Yeah, we don't want to live in a bubble. But also we had that after COVID, right, where everyone got 10 colds because they hadn't been doing anything. So it's like, oh, now we're seeing each other. I'm giving you my cold. You're giving me your cold. I'm giving you this. Like everybody got sick for like it was like an eight month period where people got sick every other month. It felt like our immune systems were like, oh, well, this is easy. I'm going to catch up on my sleep.

1:24:22Rob Reid:I'm going to stop going to the gym. Yeah. What is one thing synthetic biology might let us do that still makes the sci-fi writer in you think, wow, I can't believe we might live to see that? This is a domain of syn-bio and biology in general that is so infused with hope that, you know, conversion on wishful thinking. But really in, you know, kind of the longevity areas, there have been quite a few really interesting breakthroughs really over the last couple of years that are highly suggestive of something working potentially profoundly well. There's a lot that's going on now with personalized therapies in cancer that is accelerating.

1:25:01Rob Reid:It's not strictly due to AI. It's partly due to the acceleration in sequencing, which has kind of plateaued recently. But man, we can sequence very, very deeply at budget levels that were inconceivable several years ago. So there's a lot of exciting stuff going on in there. I do believe we'll get very good at making much more flavorful, robust, gentler on the environment food. Ah, okay. Crops and potentially even animals, not in the next two to three years, but I think we'll start seeing some really, really significant advances in less than five. There's a lot of great stuff. And like I said, I don't want to live in a world without cloud labs.

1:25:39Rob Reid:I don't want to contemplate a future without synthetic biology and all of its blessings, because they have the potential to be truly, truly extraordinary. And the thing about these things is they tend to be these compounding processes that are almost invisible for many, many, many years. Something doubles and doubles and doubles and doubles. But if it starts at the scale of a penny, you're sitting there two weeks later and it's like$14, you're pretty unimpressed. That's right. But when it gets to the end of the month, I mean, my dad had this thing and I'll never forget it. Like, would you rather have a million dollars or a penny that doubles every month?

1:26:12Rob Reid:That's right. Yeah. I was going to bring this up. You know, as the kid who took the million bucks, I was feeling pretty smart on the 21st of the month. Didn't actually get a million bucks. You know what I mean? We started doing the math, like the 21st, 22nd, ha ha, and then boom. And I think we're getting late in the month now with a lot of things in biology. And as things, because again, synthetic biology is an exponential technology, there are innate doublings that go on in there. But then there's also the interaction of things that seemed kind of unrelated, you know, like GPS and LED screens and switched telephony all would have seemed like very different disciplines.

1:26:53Rob Reid:And you would not have imagined Uber. Right. That's a good point. All of a sudden, multiple threads come together and create something or Tinder or, you know, So whatever, you know, pick your service that even the smartest people in technology 18 months before they came out would not have probably figured could be massive society transforming heads. Sure. And I think we're getting late in the month. Yeah, that's really interesting. In good and bad ways. In good and bad ways. I was going to say. Let's focus on the good right now. Hopefully we get some of the good things before we get any of these extremely bad things.

1:27:25Thank you for coming on the show again. I appreciate it, man. Thank you as always for having me. What could happen if humans mess with genes and if labs leak superbugs? In this clip, Rob Reed chats about DNA printers that could either save us or, well, not by churning out vaccines or viruses right from our living rooms.

1:27:44Rob Reid:The terrifying thing is COVID is pretty damn benign compared to what could have easily happened this time around or what could very easily happen next time around, particularly if the next bug is maliciously designed. Society produces a certain small but terrifying percentage of people every year who, for whatever reason, go to such a dark place that they become suicidal mass murderers, and their death toll is limited only by the weapons that they have. Technology is the force multiplier. The 1918 flu virus, which killed at a much, much, much greater scale than COVID, and the smallpox genome. Both of those are online and anybody could find them within a short number of minutes.

1:28:31Rob Reid:The time would soon come where somebody could take that and reanimate that. And something which scares the bejesus out of me, which is an influenza virus, not a coronavirus, is H5N1 flu. It kills 50 to 60 percent of the people that it infects. Two independent research groups, one in Holland and one in Wisconsin, took it upon themselves. and they basically made it capable of aerosolized transmission through the breath. No lab is secure enough to keep this stuff running out. And this is a pathogen that could quite literally topple civilization if it's contagious enough. If the lights shut off on a countrywide basis, after a shockingly small number of days, civilization starts to teeter and eventually topple.

1:29:20For more things that'll keep you up at night and whether we can handle the power we're playing with, check out episode 510 of The Jordan Harbinger Show. Big thank you to Rob Reed. Funnily enough, while we were recording this, the New York Times published an article stating that Anthropic, the company behind Claude and other major AI infrastructure, had discovered potential would-be terrorists using its models to learn how to create bioweapons. So this is not just fear-mongering. It's, well, justified fear-mongering. We got the push notification when we turned our phones back on. after recording the show and we were like, what timing?

1:29:54So the takeaway here isn't that we should crawl into a bunker and start boiling the mail. Synthetic biology could give us better cancer treatments, better food, longer lives, technologies that we can barely imagine right now. But as Rob puts it, we're getting late in the month on the exponential curve and the same tools that can cure disease can also make extremely dangerous capabilities accessible to a whole lot more people. So maybe we install the smoke detectors before the kitchen is on fire this time. Links and resources for Rob will be in the show notes as usual. Advertisers, deals, discount codes, and ways to support the show at jordanharbinger.com slash deals.

1:30:30Please consider supporting those who support the show. Six Minute Networking over at sixminutenetworking.com. I'm at jordanharbinger on Twitter and Instagram. You can also connect with me on LinkedIn. And this show is created in association with Podcast One. My team is Jen Harbinger, Jace Sanderson, Robert Fogarty, Tata Sidlouskis, Ian Baird, and Gabriel Mizrahi. Remember, we rise by lifting others. The fee for the show is you share it with friends when you find something useful or interesting. In fact, the greatest compliment you can give us is to share the show with those you care about. So if you know somebody who's interested in bioweapons, AI, synthetic biology, tech in general, definitely share this episode with them.

1:31:03In the meantime, I hope you apply what you hear on the show so you can live what you learn. And we'll see you next time.

1:31:15Ever notice how life's best stories don't happen in your living room?

1:31:19Rob Reid:They happen on the open road, out on the water, or parked under the stars. At Progressive, they get that you want to focus on the experience, not worry about the what-ifs. That's why they offer quality insurance designed for your ride, whether that's a boat, RV, or motorcycle. Adventure with confidence. Visit Progressive.com and see how easy it is to protect your favorite way to get away. Progressive Casualty Insurance Company and Affiliates. Not available in D.C. Prices vary based on how you buy. This episode is brought to you by Progressive Insurance. Do you ever think about switching insurance companies to see if you could save some cash?

1:31:56Progressive makes it easy to see if you could save when you bundle your home and auto policies. Try it at Progressive.com. Progressive Casualty Insurance Company and Affiliates. Potential savings will vary. Not available in all states. For a small business owner, every day is full of surprises. Some great, some not so great. Like when a client cancels their order at the last minute. But here's a surprise you will like. Progressive provides small business owners with 30 customizable coverage options to help keep their business going strong. So go ahead, surprise yourself. Get a quote in as little as eight minutes at ProgressiveCommercial.com.

1:32:32Progressive Casualty Insurance Company and Affiliates and Third Party Insurers. Coverage is not available in all states or for all vehicles and coverage selections.

From the publisher

When amateur mad scientists can leverage AI to engineer future pandemics, we need to be prepared to counter the threat. Rob Reid is here to tell us how.

Full show notes and resources can be found here: jordanharbinger.com/1390

What We Discuss with Rob Reid:

  • How a terrifying thought experiment reframes bio risk. Rob argues the scariest scenario isn't a Soviet-style industrial complex but the mere passage of time, since an exponential technology turns what a handful of elite specialists can do today into routine undergraduate coursework tomorrow.
  • Why a $100 million USAID program called DEEP VZN struck Rob as a suicide pact. The plan was to hunt 15,000 undiscovered viruses, sort them for which posed the gravest threat, then publish the findings openly. Rob and MIT's Kevin Esvelt pushed until it was quietly defanged and later killed.
  • What makes smart, well-meaning people defend this work. Rob points to an insider culture where open-science absolutism, confidence that we'd out-invent any bad actor, and vaccine promises that collapse under scrutiny combine into a stew that lets good people talk themselves into terrible ideas.
  • Why Rob keeps returning to aviation as the model. America had 40 to 50 hijackings a year until basic airport screening arrived in 1972, and the next year there were zero. You can't make a threat impossible, but a high enough bar forces conspiracies big enough for law enforcement to catch.
  • Fortunately, the countermeasures are cheap and already understood. Wastewater sequencing at major airports, stockpiled gear for frontline workers, and germicidal UV light in schools and hospitals would cost a sliver of the defense budget. Denial and despair both produce nothing.
  • And much more...

And if you're still game to support us, please leave a review here — even one sentence helps!

This Episode Is Brought To You By Our Fine Sponsors:

See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

More from The Jordan Harbinger Show

All 576 episodes
1390: Rob ReidThe Jordan Harbinger Show · 1 h 29 min
Listen in VO