Currently viewing a development environment

The lab-leak hypothesis for COVID-19 is becoming a conspiracy theory

To explain where SARS-CoV-2 came from, look at processes with scientific explanations and precedents

Dan Samorodnitsky

Senior Editor

No one knows for sure where SARS-CoV-2 came from. I don’t know, no one does. But there are two ideas. The first is the virus was harbored by an unknown animal, likely bats, where it mutated and picked up the ability to infect humans. Many pandemic viruses — Ebola, the 1918 flu — emerged this way. The other is that the virus was deliberately created in a lab at the Wuhan Institute of Virology, a research center near to where initial outbreaks were first detected in China.

Though there is no direct evidence for either idea, the natural origins hypothesis has scientific precedence. The coronavirus family of viruses, that SARS-CoV-2 is a part of, have spilled over into humans (SARS and MERS) and caused pandemics. Natural origin also takes into consideration natural phenomena that happen all the time in wild viruses — they reproduce rapidly, mutate frequently, acquire bits of DNA like a boat collecting barnacles, and change behavior, particularly when they shift from one host to another. These processes happen in all viruses.

But, the permanent uncertainty of SARS-CoV-2’s origins has made other explanations, no matter how complex, attractive. Now, the lab-leak hypothesis has taken on the rhythm and melody of conspiracy theory.

The huge “scoop” from the Wall Street Journal earlier this week is that three staffers at the Wuhan Institute for Virology had become ill in late 2019, during cold and flu season. The article quotes a Trump official who said it sounded fishy, and that’s it. There is no even indirect scientific evidence that the virus was created in a lab. Nevertheless, there seems to be some wish, some desire for China to be implicated in a cover-up.

In 2002, SARS spilled over from horseshoe bats into humans. MERS, caused by another coronavirus, spilled over from camels in 2012. Related coronaviruses within SARS-CoV-2’s subfamily have been identified in wild bats and pangolins. And, the virus's features that seem to scream out for the hand of a synthetic biologist are better explained by the kind of random, driftwood mutations that viruses pick up constantly. 

SARS-CoV-2 uses its spike protein to bind tightly to ACE2, a protein on cell surfaces in humans and other animals. But, research has shown that this binding is actually not biochemically ideal, which makes the idea that it was synthetically created less likely; this ability to bind ACE2 could have easily arisen through common routes of mutation. The ballyhooed “furin cleavage site” — a site where the viral spike protein is cut, facilitating infection into a cell — was created by an out-of-sequence insertion of a small piece of DNA resulting in yet another non-ideal biochemical reaction.

New viruses emerge all the time. Constantly. The number of viral particles on the planet is more than the number of stars in the universe; the rate that viruses reproduce, their ability to quickly mutate and adapt to new environments and new hosts, means there are functionally an infinite number of viruses on the planet.

So, figuring out where this particular virus came from will be a challenge. It can take years, decades, or more to find the source of a virus. Ebola, for instance, was identified in 1976, has caused multiple epidemics, and we still don’t really know what animal it spilled over from. To confirm beyond a reasonable doubt the virus’s origins, we’d have to sample wild animals and sequence the viruses they carry to find a close genetic relative, an astronomical task, haystacks within haystacks. In the absence of a smoking gun, there's still good research that points in one direction. Take the phylogenetic analysis in preprint this week that, once again, suggests bats as a reservoir for SARS-CoV-2, with pangolins or civets as possible intermediate steps. 

If the question is “are both hypotheses possible?” the answer is yes. Both are possible. If the question  is “are they equally likely?” the answer is absolutely not. One hypothesis requires a colossal cover-up and the silent, unswerving, leak-proof compliance of a vast network of scientists, civilians, and government officials for over a year. The other requires only for biology to behave as it always has, for a family of viruses that have done this before to do it again. The zoonotic spillover hypothesis is simple and explains everything. It’s scientific malpractice to pretend that one idea is equally as meritorious as the other. The lab-leak hypothesis is a scientific deus ex machina, a narrative shortcut that points a finger at a specific set of bad actors. I would be embarrassed to stand up in front of a room of scientists, lay out both hypotheses, and then pretend that one isn’t clearly, obviously better than the other. 

Besides the hazy science, there is an undeniable political aspect to this argument. When violence against Asian people in the US is spiking, it’s naive at best and violent gaslighting at worst to pretend that supporting an evidence-free hypothesis that clearly adds fuel to the idea that China inflicted COVID-19 upon the world, that they did this to us, is noble scientific dispassion. There’s a choice being made here between two ideas — one that falls neatly within the world of biology, and the other that knots together conspiracy theory, political intrigue, and xenophobia.

And since we will never be able to prove the exact moment that SARS-CoV-2 jumped from an animal to a human, this is instead going to devolve into a culture war. We are witnessing the real-time birth of a new axis of half-truths, convenient omissions, and quackery.

The most bothersome thing about all this is that it does not particularly matter where SARS-CoV-2 came from. Making a scapegoat out of China doesn’t do anything about the political and economic systems that allowed millions to die, especially in wealthy nations like the US that could easily afford muscular public health responses. In March, Marion Renault wrote in The New Republic:

“We have known for years that resource extraction and human expansion into wildlife habitats could lead to emergent zoonotic diseases. We have known that inadequate health coverage and sick leave policies could spread illness; countless reports and studies in recent years have chronicled the draining of public health resources, the erosion of science and public trust in it (anti-masking falls into this category), and the failures of the U.S. health care system to equitably provide affordable medical care to all.”  

Creating a webbed story of cover-ups and conspiracy allows us to ignore how in many ways all humans caused the COVID-19 catastrophe. What if it wasn’t one bad Chinese government’s fault, but the whole world’s fault for destroying habitats, mining too deeply, and creating the perfect conditions for natural viral spillovers? The lab-leak hypothesis will soon be in league with climate change denialism — a conspiracy that absolves humanity of its mistakes, and lets us live our lives as if nothing had ever happened.

Comment Peer Commentary

We ask other scientists from our Consortium to respond to articles with commentary from their expert perspective.

Karen Kemirembe


Thanks Dan for summarizing the lab-leak vs zoonotic spillover hypotheses. After reading this, I was wondering:

  1. You mention that "three staffers at the Wuhan Institute for Virology had become ill in late 2019, during cold and flu season." Do you know whether they and their close contacts were ever tested for SARS-CoV2 antibodies in 2020, after the virus had been identified?
  2. This Nature summary mentions that the first known COVID-19 patient’s parents visited a local wet market and suggests that a large array of samples from farmed, wild and domestic animal samples in 2019 and 2020 could give clues about the virus’s origin.
  3. Just like the Seattle flu study pivoted to SARS-CoV2 testing, perhaps there are Chinese samples from 2019/2020 that could be co-opted to see if the virus was circulating before it was detected in Dec 2019?

And yes, it is true that we all in some way have indirectly led to disease spread by going to work sick or encroaching on wildlife habitats; hopefully surveillance research continues to be funded at the local community level so that if any new pathogen emerges, we are not caught off guard again. As many have noted, spending millions of dollars on disease surveillance efforts is a lot cheaper than the trillions that have been spent on the COVID-19 pandemic response.

Dan Samorodnitsky responds:

Hi Karen!

  1. It’s not clear – the WSJ story doesn’t report on that, sorry to say.
  2. It’s my understanding that the idea of the wet market being the place where transmission occurred has largely been abandoned. To me the best thing to do in terms of where to look to help find where the virus came from is in sequencing studies. I only referred to one paper like that in the article, but there are plenty others. Here’s a biorXiv preprint from a few weeks ago, using sequencing and AI, that places SARS-CoV-2 in a group with bat and pangolin coronaviruses. It seems hard to ignore multiple genetics studies from independent groups using different approaches all coming to similar conclusions.
  3. I’m sure that’s the case – there have been other studies that have pinned the virus down in countries outside of China as early as December 2019. Here’s one showing it in Italy a few months before things got really bad there in early 2020.

Elise Cutts


Massachusetts Institute of Technology

I’ve been hearing more and more about the lab leak hypothesis lately, but usually not in quite the form you describe. Among the scientists (especially wet-lab biologists) I’ve talked to, there’s less suspicion that SARS-CoV-2 was “unleashed” and more musings about the possibility of some poor researcher accidentally infecting themselves and setting off the whole thing by accident. Still, while this story isn’t quite the “China did it” finger-pointing you describe, I can’t shake the feeling that it could have bubbled out of the same pot of anti-Asian racism that brewed much of the more explicitly toxic rhetoric around the possibility of a lab leak.

Distaste aside, I do still believe that there needs to be room for these two — and potentially more — competing hypotheses in the effort to understand the origin of SARS-CoV-2. As you pointed out, we don’t know what happened yet, so it’s too early to take possibilities off the table entirely. That said, there is absolutely no reason to treat both hypotheses equally. Science is under no obligation to hold all explanations as equally plausible, especially when, as in this case, one has the weight of considerable precedent behind it.

To borrow a term from astrobiology, the lab origin for SARS-CoV-2 should be considered the “hypothesis of last resort” — the conclusion we accept only when no other, more conventional explanation can satisfactorily explain the observations given the evidence at hand. Just as planetary scientists and biologists will need to painstakingly eliminate all feasible non-biological explanations for any putative biosignatures we may one day discover on Mars before crying alien, anyone involved in tracing the origins of the COVID-19 pandemic should be careful to build a convincing case against an animal origin before crying conspiracy.

Tara White responds:

while this story isn’t quite the “China did it” finger-pointing you describe, I can’t shake the feeling that it could have bubbled out of the same pot of anti-Asian racism that brewed much of the more explicitly toxic rhetoric around the possibility of a lab leak.

I thought of this as well. If there was a leak, intentional or not, it certainly won’t help alleviate this issue. I do agree that all possible origins should be investigated and whatever is learned should be used to understand and address future outbreaks.

Dan Samorodnitsky responds:

I agree, with you and Francesco, I’m not saying don’t investigate the idea. But I also feel no compunction about identifying what, to me, is a more plausible hypothesis. Isn’t that what scientists are taught to do?

At the beginning of the pandemic the hypothesis was that researchers had engineered the virus and accidentally released it. Now it seems to have shifted towards “they sampled it and accidentally released.” While that version is certainly more plausible, it’s also, to me, indistinguishable from a natural spillover hypothesis, but with the added ability to blame somebody specific for it.

Francesco Zangari responds:

Totally agree with you Dan. I hope for a fair unbiased investigation without geopolitical influences to get to the bottom of this. GOF research should not be banned as Soren adds, but if it can be made safer if in case it is found to have caused this pandemic then I am all for it.

Jayati Sharma

Genetics and Epidemiology

Johns Hopkins University

I definitely agree that the genetic evidence supporting a zoonotic spillover event is overwhelming against that of a lab leak, and that the latter continues gaining traction in an attempt to place blame elsewhere. In a world of half-read headlines and social media-based opinion making, though, I wonder how this strong, evidenced argument echoed by you and others can or will reshape public opinion on inevitable future infectious disease outbreaks. In what is increasingly becoming our post-truth society, how do scientists, or more importantly, how do evidence-based positions, re-take the reins?

Dan Samorodnitsky responds:

Honestly I wish I had an answer!

Francesco Zangari responds:

While I am not one for conspiracy, the US government appears to be taking this lab leak hypothesis quite seriously (see here: Covid: Biden orders investigation into virus origin as lab leak theory debated - BBC News) so I do think it’s worth a conversation. Regardless, it does not excuse the mishandling of the pandemic.

Simon Spichak


I think it’s really hard to separate the science from the politics with the Lab Leak hypothesis. A few countries are suspected of pushing coronavirus and vaccine misinformation to destabilize our confidence in science and the media.

  1. Russia is suspected in trying to pay science bloggers to spread Pfizer misinformation.
  2. Countries spreading misinformation to blame the West.

It leaves me wondering whether the popularity of the theory might be driven by domestic or foreign disinformation campaigns. With social media, it’s so much easier for fringe theories to spread. Since people aren’t comfortable with uncertainty and it’s hard to explain that science can’t prove something didn’t happen.

With the under-current of anti-Asian sentiment, I feel some outlets are irresponsible in the ways they write about the Lab Leak Hypothesis. It’s treated like its something that is just as likely or probable as spillover.

Soren Emerson


Vanderbilt University

Sometimes purveyors of the lab leak hypothesis point to gain-function-research as a supposed culprit in the more recent versions of the lab leak hypothesis you describe. Based on your research for these piece, I was wondering if you had tracked down where blame of gain-of-function research arose from?

My research experience is in C. elegans genetics and day to day gain of function alleles are just another genetic tool. For instance, if one suspects that a gene is responsible for a certain phenotype they could knock out the gene or enhance its function, resulting in either a loss of the phenotype or enhancement of the phenotype, respectively. In some cases knocking out a gene does not change its associated phenotype if there are other genes serving a redundant functional role so the only way to study the phenotypic output of the gene is using a gain-of-function allele.

Following the logic of those who implicate gain-of-function techniques as a culprit leads to the terminal conclusion of banning generation and use of gain-of-function alleles, which is a useless, and harmful, end for public health and basic research.

Dan Samorodnitsky responds:

Based on your research for these piece, I was wondering if you had tracked down where blame of gain-of-function research arose from?

Hey Soren, good question. I don’t know when specifically it entered into people’s lexicon. I personally find it a little weird because I’ve never heard it isolated like that, usually it has the word “mutation” hanging off the end of it. When people ask “what about gain-of-function?” it sounds to me like “what about evolution?” — the words are correct, grammatically, but the meaning isn’t really there.

My personal feeling about is it that it’s kind of like “critical race theory” — just a phrase people have latched on to as a toehold for arguing. The debate about doing gain-of-function research in potentially dangerous viruses is older than COVID, but I agree that blaming the concept itself is a terrible idea and would make lots of research basically impossible.