Where did COVID-19 come from
It does not exist / is not a new virus / 5G / aliens
Non-zoonotic origins
Transfer at a wet market
Transfer in the wild to a civilian
Transfer in the wild to a virus researcher
Accidental transfer in a lab with no modifications
Accidental transfer in a lab after research gain-of-function modifications
Planned deliberate release
Deliberate release by a rogue employee
Get Ṁ600 play money
Sort by:
bought Ṁ3 Planned deliberate r... YES

Reading Rootclaim made me realize I was silently assuming "by China" for this option, while it seems on a surface level pretty possible for a well founded external agent to synthesize a virus close enough to the ones studied there, based on published research, and release it. Is there some reason that's not a feasible (though low probability) explaination?


They would have to start with something that's very close to SARS2. And prior to 2019, there was nothing in the published literature to suggest Wuhan had viruses close to SARS2. So they would need to have access to the sequences in Wuhan through some other means.

@CodeandSolder Full sequence of Ratg13 was released in Jan 2020. Even if someone outside had access to it pre-pandemic, it is not straightforward to get to something like SARS2 using Ratg13's reported sequence.

There was a partial 98.9% match with SARS2 and an RdRp sequence at WIV but that was not the full sequence.

@Akzzz123 I'm not saying it's straightforward, but synthesis from sequence is AFAIK doable at very reasonable cost, from there on the road is the same as lab leak. And even if the full sequence wasn't published stealing a few kb of data from a non-cybersecurity-oriented facility shouldn't pose a significant obstacle to a determined actor, Stuxnet happened after all

I'm not saying it's in any way probable, just that it's not completely absurd

@CodeandSolder Yeah, if they have the sequences, cost would not be a barrier. It would be more around what mutations to add to make it a pathogen that can cause a pandemic. So there would be some experimentation involved (gain of function research), but nothing that cannot be done or is not already being done.

@Akzzz123 I would bet money on the sequences being on at least one PC a researcher accesses e-mail from and could be convinced to open grant_proposal.pdf.exe with little effort, not that a state actor couldn't easily work around that if needed

@CodeandSolder You wouldn't even need to. These idiots were planning to open-source the whole thing.


Polling manifolders on the same question

bought Ṁ0 of Transfer at a wet ma... YES

Added a bit of subsidy. Will add more once market has some more traders

It does not exist / is not a new virus / 5G / aliens

I love this one

bought Ṁ1 of Deliberate release b... YES

@JaimeSantaCruz after a few of my "funny conspiracy theories" markets resolved YES I learned not to dismiss any options lol

bought Ṁ25 of Transfer at a wet ma... YES

@CodeandSolder What do you make of those resolutions?

@JaimeSantaCruz mostly that a lot of things happen, by probability some will be weird, and it's unwise to be confident in dismissing things based on heuristics

@JaimeSantaCruz for example based on metadata I would have guessed below 50% on /CodeandSolder/is-republicans-motion-to-vacate-spe but the underlying analysis was sound and it did happen. It is infeasible to read through everything and heuristics are unavoidable but it's the lowest information form of predicting possible and should not be solely relied on if stakes exist

bought Ṁ10 of It does not exist / ... NO

@CodeandSolder Thanks for your insights!

@CodeandSolder I feel like that is also an error in the opposite direction. Forecasts on motion to vacate were moderately high even before he was elected (https://manifold.markets/octothorpe/during-the-118th-congress-will-the). It's true that if you knew nothing about the topic you might think it was more like a conspiracy theory than a real possibility, but some investigation (or just looking at the markets) quickly shows otherwise.

Being a good forecaster means being able to (on average) dismiss options in proportion to their actual likelihood.

That answer has strong "one of these is not like the others" vibe. SARS-COV-2 not being a new virus and actually having been around for a while seems quite plausible.

@IsaacKing I don't think it's that plausible, just CDC sequences ~7k influenza samples a year and SARS-COV-2 has relatively distinct symptoms, it seems it got caught very soon after the currently accepted patient zero.


SARS-CoV-2 without the furin cleavage site cannot transmit well. The furin cleavage site also makes SARS-CoV-2 more severe. Since the feature that makes SARS-CoV-2 transmit between humans also makes it severe, it's very unlikely that it was spreading undetected.

Proponents of the seafood market theory would have to reconcile this apparent contradiction that SARS-CoV-2 somehow magically learned to jump to humans (ie acquired a furin cleavage site), at least twice according to their latest dogma, at the seafood market, but left no trace of this evolution anywhere else along the road despite Wuhan being at least 1800km, by road, from where SARS-CoV-2's closest ancestors are found.

Proponents of the lab leak theory can easily explain why this evolution was detected only in Wuhan since the scientists in Wuhan had co-authored a grant proposal in 2018 that showed they wanted to introduce furin cleavage sites into novel sarslike viruses.

SARS-CoV-2 virus lacking the S1/S2 furin cleavage site was shed to lower titres from infected ferrets and was not transmitted to cohoused sentinel animals, unlike wild-type virus.


By contrast, the ΔPRRA mutant was attenuated in a human respiratory cell line and had reduced viral pathogenesis in both hamsters and K18-hACE2 transgenic mice (which express human ACE2).


You might find some of the scenarios discussed here useful (Annex A and Annex B)


There are two fundamental questions:

  1. How was the virus synthesized?

  2. How did it infect the first human(s)?

Trying to capture all the scenarios as mutually exclusive choices could get a bit messy.

Some of these answers aren't mutually exclusive, can you explain how you plan to resolve in those cases?

@IsaacKing could you give some examples?

"Accidental transfer in a lab after research gain-of-function modifications" would be "Non-zoonotic origins", wouldn't it?

@IsaacKing if the virus was not initially collected from an animal I guess that's correct, it would resolve to the more specific one, here accidental transfer