Daily-Dose

Contents

From New Yorker

From Vox

But, as Politico reported last week, USAID expects it will have to stop funding those activities in the second half of the year without an infusion of new funding from Congress. The deal announced this week was maybe the last best chance to get the money approved.

That represents a humanitarian failure. Though the official death toll in these countries is generally lower than in many wealthy countries, the statistics may dramatically undercount the actual devastation of the pandemic. A recent analysis from the World Bank suggested that the real number of pandemic deaths in Kenya could be nearly six times higher than the official count.

“Across Africa, the Covid-19 death count may be higher than official counts, pointing to the urgent need to ramp up global vaccine access,” the World Bank researchers wrote, less than two months before Congress would fail to do exactly that.

And cutting those efforts not only puts people’s health in Africa at risk, but Americans’ health as well. Experts have been warning for months that if the virus continues to circulate in other parts of the world, new variants could continue to emerge. The omicron variant, which was first detected in South Africa, killed more than 2,500 Americans every day at its peak.

On top of the risk to global health, outside estimates suggest even the funding in the new legislation dedicated to America’s Covid-19 response may not be sufficient.

The $10 billion authorized in the new legislation will be used to buy more Covid-19 medications, tests, and vaccines. But it’s not clear there will be enough funding available to sustain the vaccination campaign into the future.

Fourth shots are now okayed for every American over 50. According to a Kaiser Family Foundation analysis, the US barely has enough vaccine doses in hand to provide all four shots to 70 percent of the over-50 population. If we wanted to reach 100 percent of that cohort, the country is 225 million doses short.

For now, experts don’t believe a fourth shot is necessary for younger people unless they are immunocompromised. But that could change if a new and more dangerous variant were to become dominant — something that will be more likely if large portions of the world remain unvaccinated. And if the United States were to eventually approve the fourth dose for all ages, the country is roughly 500 million shots short of the necessary number to fully vaccinate everyone.

Acquiring the necessary doses could cost up to $10 billion all on its own, according to the estimates. Out of the $10 billion in new funding negotiated by Congress, at least $5 billion must be spent on therapeutics, according to the bill text, and another $750 million would be earmarked for research on shots targeting new variants. That would leave $4.25 billion for buying more of the current vaccines and for increasing testing capacity.

In a separate analysis, KFF’s researchers warned that future vaccine doses may prove to be harder to come by, too. To date, the US has had a sufficient vaccine supply because it bought the doses ahead of time. Vaccine manufacturers had a guaranteed market for a certain number of doses. But without pre-purchasing agreements, companies may not produce enough doses and the US would be left competing with the rest of the world for whatever supplies are still left.

“Together, this could contribute to shortages of supplies if and when the next Covid-19 wave hits and demand increases,” the researchers wrote.

Rather than make investments now that could pay off when Covid-19 inevitably surges again, Congress is cutting corners. The cost of this myopia could end up being felt all over the world in the months to come.

Schoenbrun and I are around the same age. They know all about art on the Internet, both as a filmmaker and as a director of underground-favorite series The Eyeslicer, which curates ephemera into kind of a new-millennium variety show. In addition to We’re All Going to the World’s Fair, their 2018 film A Self-Induced Hallucination, about the Slenderman phenomenon, is playing in the series, as is a work-in-progress screening of their (terrific) film Girl Internet Show: A Kati Kelli Mixtape, which compiles the daring work of an experimental video artist. We’re All Going to the World’s Fair captures the experience of the internet through the eyes of an isolated teenager; the other films use footage collected from the internet to interrogate an important question. As Schoenbrun put it to me: “Why do we go to the internet to narrativize ourselves?”

It’s a tough one. When I asked for Schoenbrun’s answer, they were swift to reply: “Loneliness is the one-word answer. A desire for connection, a desire for meaning, that desire for some sort of narrative that will make us feel that our lives are moving toward something that matters.” At its best, Schoenbrun went on, the internet affords the opportunity to make that connection in a place where it’s easier and even encouraged to engage in identity play, to try on different identities for size. But alongside those connections and that exploration is a darker side, they said. “When the internet is scaring me, it feels more like propaganda.”

The program Schoenbrun put together to contextualize their own films starts with the disturbingly prophetic vision of Videodrome, but then keeps going, with movies that are almost entirely available to stream or digitally rent at home. (An innovation, ironically enough, that’s available to us only because of the internet.)

A 
computer desktop with many windows open, including images of several teenagers who look frightened. Universal Pictures
A scene from Levan Gabriadze’s Unfriended.

Watching the series in order feels a bit like taking a tour through my own psychic development. From Videodrome, we jump to the mid-90s for the punk-inflected paranoia of Hackers (1995) and The Matrix (1999), which foresaw the internet as a tool to control the unwitting masses. Chris Marker’s Level Five (1997) is a “fictional documentary” that, through the eyes of a grieving woman, acts like a surreal tour through early digital culture. Then there are two films from 2001 (the year I started my undergraduate studies in information technology and computer science, as it happens): Pulse, in which ghosts travel through the web, and All About Lily Chou-Chou, in which a pop star’s teen fans connect anonymously on the oh-so-familiar web bulletin boards.

And human connection can be horrifying and threatening. The 2014 film Unfriended, which takes place entirely on the laptop of a teen, uses tech that already feels a little quaint (Facebook and Skype) but spins a terrifying horror story about the ways the internet allows us new places to be horrible to one another. Two 2016 films, The Human Surge and Nerve, are very different from one another — one an experimental film, the other a more conventional thriller — but they both tap into the restless loneliness and boredom of millennials and the ways the internet can be the site for connection, not always in safe or productive ways. Penny Lane’s 2018 found-footage documentary The Pain of Others is a disturbing dip into the internet of conspiracy theories via YouTubers wildly popular for their channels exploring Morgellons disease, while Chloé Galibert-Laîné’s outstanding short essay film Watching the Pain of Others interrogates her discomfort watching Lane’s film.

More recently, the hallucinatory 2020 documentary Crestone follows a group of director Marnie Ellen Hertzler’s friends, SoundCloud rappers who live a kind of post-apocalyptic life alone in a Colorado desert. Hertzler’s short film Hi I Need to Be Loved (2018) has the same kind of dreamy aesthetic, as actors read lines from spam emails that Hertzler has received. The effect of Hertzler’s work is the feeling of having your brain rewired, of the real and the unreal fusing in a way that suggests a continuum between the digital and the real.

Watching the series, I found myself unnerved and astounded, as if I were watching centuries of human experience pass by in the space of a few decades. Schoenbrun’s statement, that we go to the internet as an answer to our loneliness, rang in my ears. Overcoming loneliness wasn’t just about connection; after all, there has to be a reason the internet has evolved toward video and images, rather than remaining a place to chat with strangers in text.

When I asked Lane about this, she said that her hours and hours of watching footage from YouTubers convinced her that what they needed most wasn’t just to connect with others. It was the “very basic human need to be seen,” she said, to be recognized. To be seen for who we are. “This need for recognition is so intense and so real,” she continued. “It can be really hard to be seen in real life.”

Moments after we hung up, she texted me: “There’s a difference between ‘wanting attention’ and ‘needing recognition.’” And the internet has afforded everyone the possibility of both.

A young woman stands in front of a computer, in shadows. Toho
A scene from Kiyoshi Kurosawa’s Pulse.

Watching the internet evolve on my TV screen, I thought about the strangeness of the past two years, in which life-and-death events — millions of lives wiped out by an invisible and deadly virus, uprisings and cultural battles, wars and rumors of wars — have been largely experiences we had on the internet. I can’t imagine living through the pandemic without the internet, and yet, I’m pretty sure many of us have had our brains warped and mashed by the very tool that sometimes felt like it was keeping us alive.

So I’d add two of my own entries to Schoenbrun’s excellent list. One is Strasbourg 1518, a short film created by Under the Skin director Jonathan Glazer in the early days of the pandemic and released to the web by A24. Strasbourg 1518, as I wrote in August 2020, is “a bit of performance art that captures the frustrations of being a physical body trapped by a pandemic, but in the way only artists of the 21st century could pull off” — that is, dancers around Europe filmed themselves performing in empty rooms alone to the same piece of Mica Levi’s frenetic music, thus collaborating in isolation.

What I didn’t anticipate in August 2020, not quite yet, was how profoundly correct the film’s invocation of the mysterious and famous “dancing plague” would feel. That mass, memeable psychosis, which went on for months in medieval Strasbourg, would grow only more literal as time went on. Now, it sometimes feels as though everything going on offline is merely an accessory to the real world online. Everyone giving their takes. Everyone repeating to one another what they’ve already heard elsewhere.

And that’s why I’d add Bo Burnham’s film Inside, which I’ve watched several times since its 2021 release, obsessed and unnerved with how it captures the wild, uncontrollable thing that is today’s internet: its deranged discourses, its glut of images, its scroll-scroll-scroll addictive quality that lights up parts of the brain our ancestors didn’t even know they had. The song “Welcome to the Internet,” with its chorus “A little bit of everything all of the time,” feels like a cackle of grinning despair, which indeed is how Burnham performs it. A never-ending merry-go-round that keeps going faster and faster, and no matter how sick it’s making you, you can’t quite get off because it’s in you now.

Near the end, Burnham reminds his audience of simpler days. “Not very long ago / Just before your time / Right before the towers fell, circa ’99,” he sings, the internet was very different, with “catalogs / travel blogs / a chat room or two.” That’s the internet the movies captured, but forward-thinking directors knew there was more going on — ghosts, hackers, surveillance, and, as Videodrome foretold, our eventual merger with the thing itself.

“Honey, how you grew,” Burnham sings. “And if we stick together / Who knows what we’ll do.”


How to watch the Internet at home, thanks to … the internet

 Hector Retamal/AFP via Getty Images
Security guards line the road in front of the Wuhan Institute of Virology in China’s central Hubei province on February 3, 2021, as members of the World Health Organization team visit to investigate the origins of the Covid-19 coronavirus.

Whatever chain of events caused Covid-19, we already know that infectious disease outbreaks can originate in a lab. In 1978, a year after the final reported cases of smallpox in the wild, a lab leak caused an outbreak in the UK. Photographer Janet Parker died, while her mother got a mild case and recovered; more than 500 people who’d been exposed were vaccinated. (Smallpox vaccination can protect against smallpox even after an exposure.) Only that quick, large-scale response prevented what could have been a full-blown recurrence of the once- extinct disease.

That’s not our only close brush with the return of smallpox, a disease that killed an estimated 300 million people in the 20th century alone. Six unsecured smallpox vials were discovered sitting in a refrigerator in the US National Institutes of Health (NIH) in 2014, having been forgotten there for decades among 327 vials of various diseases and other substances. One of the vials had been compromised, the FDA found — thankfully not one of the ones containing smallpox or another deadly disease.

Other diseases have been at the center of similar lab mishaps. In March 2014, a Centers for Disease Control (CDC) researcher in Atlanta accidentally contaminated a vial of a fairly harmless bird flu with a far deadlier strain. The contaminated virus was then shipped to at least two different agricultural labs. One noticed the mistake when their chickens sickened and died, while the other was not notified for more than a month.

The mistake was communicated to CDC leadership only when the CDC conducted an extensive investigation in the aftermath of a different mistake — the potential exposure of 75 federal employees to live anthrax, after a lab that was supposed to inactivate the anthrax samples accidentally prepared activated ones.

After SARS emerged in 2003, there were six separate incidents of SARS infections resulting from lab leaks. Meanwhile, last December, a researcher in Taiwan caught Covid-19 at a moment when the island had been successfully suppressing outbreaks, going without a domestic case for more than a month. Retracing her steps, Taiwan authorities suspected she’d caught the virus from a bite by an infected mouse in a high-security biology lab.

“The fact is that laboratory accidents are not rare in life sciences,” former Senator Joe Lieberman told the bipartisan Commission on Biodefense this March. “As countries throughout the world build additional laboratories to conduct research on highly infectious and deadly pathogens, it’s clear that the pace of laboratory accidents will naturally increase.

According to research published last year by King’s College London biosecurity researchers Gregory Koblentz and Filippa Lentzos, there are now nearly 60 labs classified as BSL-4 — the highest biosecurity rating, for labs authorized to carry out work with the most dangerous pathogens — either in operation, under construction, or planned in 23 different countries. At least 20 of those labs have been built in the last decade, and more than 75 percent are located in urban centers where a lab escape could quickly spread.

Alongside the near certainty that there will be more lab escapes in the future, engineering the viruses that could conceivably cause a pandemic if they escaped is getting cheaper and easier. That means it’s now possible for a single lab or small group to conceivably cause mass destruction across the whole world, either deliberately or by accident.

“Potential large-scale effects of attempted bioterrorism have been mitigated in the past by terrorists’ lack of expertise, and the inherent challenge of using biotechnology to make and release dangerous pathogens. Now, as people gain greater access to this technology and it becomes easier to use, the challenge is easing,” Pavel argues. The result? “Incidents of bioterrorism soon will become more prevalent.”

Dangerous research and how to combat it

The BWC, which went into force in 1975, was the first international treaty to ban the production of an entire category of weapons of mass destruction.

Identifying or creating new bioweapons was made illegal for the 183 nations that are party to the treaty. The treaty also required nations to destroy or make peaceful use of any existing bioweapons. As then-President Richard Nixon put it in 1969 when he announced the US would abandon any offensive bioweapons work of its own, “Mankind already carries in its own hands too many of the seeds of its own destruction.”

 Department of Defense/AP

Technicians work in a laboratory at Fort Detrick in Frederick, Maryland, in the late 1960s under the offensive biological weapons program the Army ran there from 1943 to 1969.

But the BWC is underfunded and little-prioritized despite the magnitude of the threat biological weapons pose. It has just a few staff members running its implementation support unit, compared to hundreds at the Chemical Weapons Convention, and a budget smaller than that of the average McDonald’s franchise. The US could easily bolster the BWC significantly with a relatively small funding commitment, and should absolutely do so.

And despite the treaty’s broad aims, much of the work to identify dangerous pathogens that could potentially act as bioweapons is still ongoing — not as part of Cold War-era covert programs deliberately designed to create pathogens for military purposes, but through well-intentioned programs to study and learn about viruses that have the potential to cause the next pandemic. That means the Biological Weapons Convention does little to constrain much of the research that now poses the greatest risk of future biological weapons use, even if the release of those viruses would be entirely inadvertent.

One such type of science is what’s called “gain of function” research, in which researchers make viruses more transmissible or more deadly in humans as part of studying how those viruses might evolve in the wild.

“I first heard about gain of function research in the 1990s, only then we had a different term for it: biological weapons research and development,” Andy Weber, former assistant secretary of defense for nuclear, chemical & biological defense programs in the Obama administration and now a senior fellow at the Council on Strategic Risks, told me. “The intent is 180 degrees off — NIH is trying to save the world from pandemics — but the content is almost entirely overlapping.”

The status of gain of function research has been hotly contested over the last decade. In 2014, after the series of scary lab safety and containment failures I outlined above and after revelations of alarming gain of function work on bird flu, the NIH, which funds much of the cutting-edge biology research worldwide, imposed a moratorium on gain of function work on pathogens with pandemic potential like influenza or SARS. But in 2017, the moratorium was lifted without much explanation.

Right now, the US is funding gain of function work at a few select laboratories, despite the objections of many leading biologists who argue that the very limited benefits of this work aren’t worth the costs. In 2021, a bill was introduced to prohibit federal research grants that fund the gain of function research on SARS, MERS, and influenzas.

Beyond the risk that a virus strengthened through gain of function work might accidentally escape and trigger a larger outbreak — which is one theory, albeit unproven, for how Covid-19 began — it can be hard to differentiate legitimate if risky research from deliberate efforts to create malign pathogens. “Because of our government support for this risky gain of function research, we’ve created the perfect cover for countries that want to do biological weapons research,” Weber told me.

The No. 1 thing he’d recommend to prevent the next pandemic? “Ending government funding for risky research that plausibly could have caused this and future pandemics.”

Another potentially risky area of virology research involves identifying animal species that act as reservoirs of viruses that have the potential to cross over into humans and cause a pandemic. Scientists involved in this work go out to remote areas to take samples of those pathogens with dangerous potential, bring them back to the lab, and determine whether they might be able to infect human cells. This is precisely what researchers at the WIV apparently did in the years leading up to Covid-19 as they searched for the animal source of the original SARS virus.

Such work was advertised as a way to prevent pandemic-capable pathogens from crossing over into humans, but it was largely useless when it came time to fight SARS-CoV-2, Weber says. “After having done this work for 15 years, I think there’s little to show for it,” Weber told me. That’s not the only view within the virology community, but it’s not a rare one. Weber thinks Covid-19 should lead to a rethinking. “As the intelligence community concluded, it’s plausible that it actually caused this pandemic. It was of zero help in preventing this pandemic or even predicting this pandemic.”

 Johannes Eisele/AFP via Getty Images
Workers are seen next to a cage with mice, at right, inside the P4 (BSL-4) laboratory in Wuhan on February 23, 2017. The P4 epidemiological laboratory was built in cooperation with French bio-industrial firm Institut Merieux and the Chinese Academy of Sciences.

There’s certainly a place for work identifying viruses at the wildlife-human boundary and preventing spillover, but the limited track record of viral discovery work has many experts questioning whether our current approach to viral discovery is a good idea. They argue that the benefits have been overstated while the potential harms have been undercounted.

At every stage of the process, such research generates the possibility of causing the animal-human spillover that the scientists intend to study and prevent. And the end result — a detailed list of all of the pathogens that researchers have identified as incredibly dangerous if released — is a gift to biological weapons programs or to terrorists.

Thanks to improvements in DNA synthesis technology, once you have the digital RNA sequence for a virus, it’s relatively straightforward to print the sequence and create your own copy of the virus (more on this below). These days, “there is no line between identifying a thing as pandemic capable and it becoming available as a weapon,” Esvelt told me.

The good news? It shouldn’t be hard for policymakers to change course on dangerous research.

The NIH funds a large share of biology research globally, and a renewed NIH ban on funding dangerous research would significantly reduce how much of that dangerous work takes place. If the US adopts firm and transparent policies against funding research into making pathogens deadlier or identifying pandemic-capable pathogens, it will be easier to exercise the global leadership needed to discourage that work in other countries.

“China funds this research too,” Esvelt told me. It might be that, spooked by Covid-19, they’re open to reconsidering, but “if we don’t stop, it’s going to be really hard to talk to China and get them to stop.”

All of that amounts to a simple prescription for policymakers: Stop funding dangerous research, and then build the scientific and policy consensus necessary to get other nations to also stop funding such research.

Behind that simple prescription lies a great deal of complexity. Many discussions of whether the US should be funding dangerous research have run aground in technical arguments over what counts as “gain of function” work — as if the important thing is scientific terminology, not whether such research might trigger a pandemic that could kill millions of people.

“94% of countries have no national-level oversight measures for dual-use research, which includes national laws or regulation on oversight, an agency responsible for the oversight, or evidence of a national assessment of dual-use research,” a 2021 report by the Johns Hopkins Center for Health Security and Nuclear Threat Initiative found.

And if that were to happen, the result could be as bad or worse than anything nature can cook up. That’s precisely what happens in a pandemic simulation put on in 2018 by the Johns Hopkins Center for Health Security. In the fictional scenario, a terror group modeled on Aum Shinrikyo engineers a virus that combines the high transmissibility of parainfluenza — a family of viruses that generally cause mild symptoms in young children — with the extreme virulence of the Nipah virus. The result is a supervirus that in the exercise eventually kills 150 million people around the world.

DNA synthesis and how it changes the bioweapons calculus

“Advances in synthetic biology and biotechnology make it easier than ever before to make pathogens more lethal and transmissible, and advances in the life sciences are occurring at a pace that governments have been unable to keep up with, which increases the risk of deliberate or accidental releases of dangerous pathogens,” Lieberman told the bipartisan Commission on Biodefense in March.

One of the most exciting recent areas of progress in biology has been the increasing ease of DNA synthesis — the ability to “print” DNA (or RNA, which makes up the genetic material of viruses like influenzas, coronaviruses, measles, or polio) from a known sequence. It used to be that creating a specifically desired DNA sequence was incredibly expensive or impossible; now, it is much more straightforward and relatively cheap, with multiple companies in the business of providing mail-order genes. While scientific skill is still very much required to produce a virus, it is nowhere near as expensive as it used to be, and can be done by a much smaller team.

 Eric Piermont/AFP via Getty Images
Thomas Ybert, co-founder and CEO of biotech company DNA Script, works on the beta version of the world’s first benchtop DNA printer on August 28, 2020, in Kremlin-Bicetre, near Paris. In Junew 2021 DNA script announced the commercial launch of the “printer,” which has the ability to create from scratch synthetic DNA fragments that can be used by laboratories.

This is great news; DNA synthesis enables a great deal of important and valuable biology research. But progress in DNA synthesis has been so fast that coordination against dangerous actors who might misuse it has lagged.

Furthermore, checking the sequence against a list of known dangerous sequences requires researchers to maintain a list of known dangerous sequences — which is itself something bad actors could use to cause harm. The result is an “information hazard,” what the existential risk scholar Nick Bostrom defines as “risks that arise from the dissemination or the potential dissemination of true information that may cause harm or enable some agent to cause harm.”

“DNA is an inherently dual-use technology,” James Diggans, who works on biosecurity at the industry-leading synthetic DNA provider Twist Bioscience, told me in 2020. What that means is DNA synthesis makes fundamental biology research and lifesaving drug development go faster, but it can also be used to do research that can be potentially deadly for humanity.

That’s the conundrum that biosecurity researchers — in industry, in academia, and in the government — are faced with today: trying to figure out how to make DNA synthesis faster and cheaper for its many beneficial uses while ensuring every printed sequence is screened and hazards are appropriately handled.

If that sounds like a challenging problem now, it’s only likely to get worse in the future. As DNA synthesis gets ever cheaper and easier, many researchers anticipate the creation of tabletop synthesizers that would allow labs to simply print their own DNA as needed for their research, no middleman needed. Something like a tabletop synthesizer could make for awesome progress in biology — and worsen the challenge of preventing bad actors from printing out dangerous viruses.

Furthermore, as DNA synthesis gets cheaper, screening for dangerous sequences becomes a larger percentage of the cost, and so the financial advantage of cutting corners on screening could become bigger, as companies that don’t do screening may be able to offer considerably lower prices.

Esvelt and the team he works with — which includes US, EU, and Chinese researchers — have developed a framework for a potential solution. They want to maintain a database with hashes of deadly and dangerous sequences — mathematically generated strings that correspond uniquely to each sequence but can’t be reverse-engineered to learn the dangerous original sequence if you don’t already know it. That will allow checking sequences against a list of deadly ones without risking anyone’s privacy and intellectual property, and without maintaining a public list of deadly sequences that a terror group or bioweapons program could use as a shopping list.

“Later this year, we anticipate making DNA synthesis screening available for free to countries worldwide,” Esvelt told me.

To make things truly safe, such a proposal should be accompanied by government requirements that DNA synthesis companies send sequences on for screening against a certified database of dangerous sequences like Esvelt’s. But the hope is that such regulations will be welcomed if screening is secure, transparent, and free of charge to consumers — and that way, research can be made safer without slowing down progress on legitimate biology work.

International governance is always a difficult balancing act, and for many of these questions we’re going to need to keep revisiting our answers as we invent and improve new technologies. But we can’t afford to wait. The omicron variant of Covid-19 infected tens of millions of people in the US in the space of just a few months. When a disease hits, it can hit fast, and it can be too late by the time we know we have a problem.

Thankfully, the risk of a serious catastrophe can be much reduced by our choices in advance, from screening programs to making deadly viruses harder to engineer to global efforts to end research into developing dangerous new diseases. But we have to actually take those steps, immediately and on a global basis, or all the planning in the world won’t save us.

From The Hindu: Sports

From The Hindu: National News

From BBC: Europe

From Ars Technica

From Jokes Subreddit