How do we respond to risks past, present and future?
The probability of winning the PowerBall jackpot is one in 292,201,338; the probability that Earth and the 2024 YR4 asteroid will collide on December 22, 2032 is 1.9% according to the most recent research, or odds of roughly one in 52 (up from the earlier estimate of one in 76 by the International Asteroid Warning Network and NASA’s Jet Propulsion Laboratory).
Most risks we confront are low probability and highly stochastic. What’s the chance that a cargo ship will block the Suez Canal? Or that a critical piece of security software like CrowdStrike will cause a worldwide computer outage? Even though the answer is “quite low” for both, there are millions of other hypothetical ways in which business can be temporarily halted. Past is not prologue for the future — the chance that another cargo ship blocks the Suez or CrowdStrike turns off the global economy again is presumably lower than ever. Our solution then is insurance: rather than predicting which of these crises may come to pass, we protect ourselves by pre-paying premiums in exchange for a payout when business is interrupted.
Yet an asteroid strike to Earth with odds of one to 52 feels like something else entirely. Earth may not have an insurance mechanism (Muskian visions of occupying Mars aside), yet we aren’t powerless in the face of impending doom. With echoes of the films Armageddon and Deep Impact, it feels like a crisis that humanity can proactively avoid with the right cooperation and intellect (to that end, we’re already exploring how to fire nuclear weapons at this particular rock to stop the apocalypse).
Why can we cooperate, even potentially among global adversaries? The legibility and tangibility of the risk help tremendously. We understand physics (although it’s worrisome that even just a few years out, scientists still have to place odds on celestial mechanics). We understand that if a big rock pierces the atmosphere, large-scale damage is inevitable — it’s only a question of how much and where. We have knowledge of past asteroid strikes — dinosaurs are believed to have gone extinct from the impact of an asteroid near the Yucatan Peninsula in Mexico that left the Chicxulub Crater. We can also visualize future destruction and make the calculation that early action can prevent immense pain.
We see similar cooperation in areas of on-going suffering where past, present and future come together. Take renal failure: in the United States alone, about 90,000 people are waiting for a donor kidney as they spend hours each week attached to dialysis machines. Tens of thousands of Americans have already died waiting for a donor kidney, thousands on the list today will not live long enough to see an organ arrive, and the causes of renal failure aren’t going away anytime in the future.
Given the scale of suffering and on-going risk, there’s large investment to solve the problem. Take eGenesis, one of my personal favorite Lux portfolio companies. Yesterday, the company announced its second successful xenotransplantation of a porcine kidney. The organ was given to Tim Andrews, who had long been waiting for a kidney donation and was authorized to receive an experimental xeno-organ at Massachusetts General Hospital under an FDA-approved new clinical trial that can scale up to three patients. From The New York Times article:
“When I got out of the recovery room and went to the intensive care unit, I actually tap-danced between the table and my bed,” Mr. Andrews said in an interview on Thursday. “I’m so happy, it’s unbelievable.”
The company published the results of its first patient in The New England Journal of Medicine this week as well.
Similar to confronting an asteroid strike, we have the sense that it should be possible to find solutions for most diseases. Sometimes that’s through a direct treatment, such as a drug or a surgical operation like a xenotransplantation. Some treatments are less miraculous, such as switching diets or changing lifestyle factors. Nevertheless, there’s a legible cause and effect relationship between risks and solutions that drives attention and investment.
Not all risks offer the same direct relationship. Laurence Pevsner and I were booking tickets to Washington’s Reagan National Airport last week on Wednesday evening for a Riskgaming session and just as the flight options loaded in my browser, I received a push notification that AA Flight 5342 had collided with a military helicopter over the Potomac, which we now know killed all 67 people on board and was the first airline crash in the United States in 16 years.
Everyone working in aviation is deeply committed to safety, and vast investments are constantly made to prevent accidents. Even so, tragedy can still befall the system. One reason, straight from Charles Perrow, is what he dubbed “normal accidents” — highly-coupled and highly-complex systems must inevitably fail given the emergent interactions of so many parts. That’s why close-calls are a regular occurrence in the hazardously congested airspace around Reagan, and have been for years.
The deeper challenge is that cause and effect in air safety is nonlinear, and therefore, many “solutions” may actually make the problem worse instead of better. For instance, Reagan’s main runway is the busiest in the United States, and the airport handles roughly 25,000 aircraft operations a month. The reflexive answer is to cut down on these figures and the number of helicopter flights taking place over the Potomac.
Yet, consider the tragic crash of Jeju Air Flight 2216 back at the tail end of 2024. Muan Airport in South Korea handled roughly 1,500 aircraft operations in all of 2023, or just about five daily. A preliminary investigation has identified duck DNA in both engines of the plane, tentatively pointing to a bird strike as the cause for the country’s worst-ever air disaster. Would better wildlife mitigation at the airport have improved the odds? Would an investment in such a system make sense when there are only five flights per day?
Which way to go? Reduce the number of flights, and you have bored flight controllers and more inexperienced pilots alongside a lack of investment in safety systems. Add too many flights, and you have frantic flight controllers, stressed pilots and safety features that can’t handle the congestion. Even the Goldilocks middle may have its own peculiar triggers of risk that aren’t present at the extremes. Our asteroid strike is so much more linear even as its orbit curves through space.
The past safety performance of American aviation has been exemplary. Sixteen years is a long time, particularly given growing rates of air travel. We know that a disaster just happened, but does that portend more air crashes in the future? Does it even make sense to change a system that was working until it was not? One reason for the conservatism of the FAA when it comes to its systems is precisely that the systems mostly work, day in and day out.
Since we’re talking about a bird strike, let’s move on to a final risk: the specter of avian influenza. Strains of H5N1 have been circulating for more than two years in the United States, triggering widespread culling of poultry totaling more than 100 million birds. While a handful of people have died from the virus, concerns about human-to-human transmission have remained minimal.
This week offered a new surprise, as we saw one of the first outbreaks of avian flu among dairy cows. As quoted in The New York Times:
“This is not what anyone wanted to see,” said Louise Moncla, an evolutionary biologist who studies avian influenza at the University of Pennsylvania. “We need to now consider the possibility that cows are more broadly susceptible to these viruses than we initially thought.”
The preventative killing of chickens has already caused egg prices to skyrocket — what would happen to milk and beef prices if we needed to cull herds of dairy cows and steer as well? The risk of further bovine spread is very hard to mitigate given the virulence of H5N1 and the ease of its decentralized transmission. There are very few tools that can scope the problem and limit its damage, and most solutions we have, including cullings, cause new problems like grocery inflation. Given that each strain of the flu is different, past solutions may not work as well today.
We all want the risk certainty of an asteroid strike, or the agency that finding and solving disease affords. But for the vast majority of the most dangerous risks out there — pandemics, global climate disruption, technological safety, war and more — we don’t have past precedent, our solutions today are inchoate and the future is profoundly ambiguous. If only every challenge were as simple as nuking a rock.
Podcast: How Russia is bringing the cost of global sabotage to zero
This week, I sat down with political scientist Daniela Richterova, co-author of the recent report “Russian Sabotage in the Gig-Economy Era.” We discussed Moscow’s doctrine of sabotage; why the country is recruiting operatives like you hail Uber drivers; and how the United States can respond.
🔊 Listen to “How Russia is bringing the cost of global sabotage to zero”
We have posted the full transcript online, but here’s an edited and condensed extract:
Danny Crichton: In your recent paper, you noted a structural change in how Russia is managing the agents who conduct its sabotage acts. Rewind back to the seventies and eighties, and you had well-trained KGB operatives who may have had years of experience, learned multiple languages, were building up a track record. They would get something akin to an apprenticeship to learn their trade craft over a long period of time. That model seems to be changing. What's happening?
Daniela Richterova: That's exactly right. We're seeing a shift from well-trained and tested agents executioners to amateurs being recruited from all walks of life and from all sorts of countries. There was one case from Latin America. Several Eastern Europeans have been recruited too, and so have Western and other nationals. From what we have seen, this recruitment happens online. That's where the gig economy idea comes from. On platforms such as Telegram, individuals can sign up for a job. They are told how much the job would pay, and where it would take place. They're not always told what the purpose is. They can bid for a job as if they were an Uber driver. The main shift we're seeing is that these guys are not well-trained. They often don't know what they're doing. And even if they are told what they're doing, they might not really know what it takes to do it. I would argue that's also why we've seen so many arrests.
As for why this is happening, one reason appears to be that this is a cost-saving exercise. We've seen that the people who are hired aren't being paid thousands and thousands of dollars. It’s often a couple of hundred dollars paid in cryptocurrency. This model introduces quite a lot of flexibility, and it allows for a change in scale. In the past, it would've taken a long time to train someone to make sure they understand the operational environment. I found cases in archives where they talked about agents executioners and how they would be trained in whatever method they would be using to conduct sabotage operations, be it chemical attacks, be it arson, be it explosions.
We don't see any of this training with these gig economy saboteurs here. And that's why I think we've seen quite a lot of amateurish work. For instance, the attack on the bus depot in Prague before the summer — this was allegedly carried out by someone from Latin America who didn't know the environment very well, and just out of the blue started trying to set a bunch of buses on fire.
The last thing that I'll mention is that the gig economy model increases deniability. We'll see how this plays out in the long run, because we'll see whether governments investigating the agent saboteurs who have been caught recently will be able to get into the suspects’ mobile phones and get into their accounts to follow the breadcrumbs all the way to Moscow. I'm not entirely sure, but it seems likely this online gig economy model makes sabotage more deniable.
Lux Recommends
- While we are on the subject of asteroids hitting Earth, David Yang pointed out this article from Alexandra Witze at Nature on “Asteroid fragments upend theory of how life on Earth bloomed.” “Not only does Bennu contain all 5 of the nucleobases that form DNA and RNA on Earth and 14 of the 20 amino acids found in known proteins, the asteroid’s amino acids hold a surprise. On Earth, amino acids in living organisms predominantly have a ‘left-handed’ chemical structure. Bennu, however, contains nearly equal amounts of these structures and their ‘right-handed’, mirror-image forms, calling into question scientists’ hypothesis that asteroids similar to this one might have seeded life on Earth.”
- Our scientist-in-residence Sam Arbesman enjoyed the research by Sofia Sheikh of The SETI Institute asking “Could a twin Earth detect Earth?” “The findings revealed that radio signals, such as planetary radar emissions from the former Arecibo Observatory, are Earth’s most detectable technosignatures, potentially visible from up to 12,000 light-years away.”
- Laurence enjoyed Jennifer Pahlka’s post on “Jed,” a long-time civil servant holding up ancient government systems. “The system he started building in the 1980s and maintained for decades before my friend came along was called VACOLS, the Veterans Appeals Control and Locator System. It started out serving just 400 users, and expanded to 17,000 across multiple parts of the VA system.”
- Anything by Ted Chiang is worth a read, and here he is being interviewed in The Los Angeles Review of Books. “As for the impact on artists, I’d say the primary effect of AI tools is that they encourage the idea that art is no different from tightening bolts. Artists have always had to deal with commercial considerations, but it’s probably a more pressing issue now than ever before. The impulse to view everything in terms of efficiency, of reducing costs and maximizing output, is radically overapplied in the modern world. There are certain situations in which that is an appropriate framing, but art cannot be understood that way.”
- Finally, Sam recommends this retrospective with Maxis founder Will Wright and others on the 25th anniversary launch of The Sims, still one of the most successful video game franchises of all time. “The Sims was a satirical take on American consumerism. Wright echoed the grandiose claims of postwar advertising in the game’s furniture catalog, which offered Sims toasters and chairs that promised to change their lives. There was often a correlation between the price paid for an item and how much it would improve a Sim’s mood.”
That’s it, folks. Have questions, comments, or ideas? This newsletter is sent from my email, so you can just click reply.
