ClaimsEliezer Yudkowsky

“We had a limited time in the summer to rapidly test vaccines using the many many volunteers for human challenge trials, and we squandered it”

Eliezer Yudkowsky, posting on X 27 October 2020 Post by @ESYudkowsky on X, 2020-10-27 (X)

Factual Biosecurity

In summer 2020 there was a limited window to rapidly test COVID-19 vaccines using many volunteers for human challenge trials, and it was not used.

Unresolvable The public record could not settle it.

Factual: Statements about the present or the past, checked against the record: did they hold up?

The second judge would not resolve this claim; the verdict shown is the first judge's.

What would count as true or false

Judge as of 2020-10-27. True: no human challenge trial of a COVID-19 vaccine began in summer 2020, even though many volunteers (e.g. tens of thousands signed up through 1Day Sooner) had come forward, and contemporaneous sources describe summer 2020 as a period when such trials would have been feasible or useful (e.g. before conventional trials read out). Mostly true: no challenge trials ran, but the claimed 'limited window' is debatable or a detail is off. Mixed: challenge trials were partly pursued (e.g. formally approved or under way in summer 2020), or experts strongly disputed that they could have sped up vaccine testing then. Mostly false: challenge trials did start in summer 2020 but were small. False: challenge trials were carried out at scale in summer 2020.

Set before the research began.

Evidence (4)

  1. E1 Expert partnership to explore and establish Human Challenge studies of COVID-19 in the UK UK Government (gov.uk) · 20 October 2020

    “If approved by regulators and the ethics committee, the studies would start in January with results expected by May 2021.”
  2. E2 Scientists ponder human challenge trials for COVID-19 vaccines Fred Hutchinson Cancer Center · 18 June 2020

    “and I am afraid that currently there is just too much uncertainty about them to make an informed decision,”
  3. E3 Covid-19 vaccines: Should we allow human challenge studies to infect healthy volunteers with SARS-CoV-2? BMJ · 9 November 2020

    “In the current pandemic, over 38 000 people have signalled their willingness to participate in covid-19 human challenge studies with 1Day Sooner, a non-profit organisation advocating for these volunteers.”
  4. E4 Covid-19 vaccines: Should we allow human challenge studies to infect healthy volunteers with SARS-CoV-2? BMJ · 9 November 2020

    “The scientific and social value of challenge studies is inadequate, as we are likely to have one or more vaccines in hand without them.”

Reasoning

The evidence shows that the UK's COVID-19 challenge studies had not started by October 2020 and were planned for January 2021 if approved [E1], and that over 38,000 people had volunteered through 1Day Sooner [E3]. None of the evidence items shows whether any COVID-19 challenge trial began anywhere else in summer 2020, which the criteria require for this verdict.

Cite this claim

Eliezer Yudkowsky, “In summer 2020 there was a limited window to rapidly test COVID-19 vaccines using many volunteers for human challenge trials, and it was not used.” (said 27 October 2020). Verdict: unresolvable. Reliability API, claim c_e990286a0730, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_e990286a0730

This claim is in dataset release v1.1.2, with the same verdict.

Reply to this claim

For Eliezer Yudkowsky, or someone speaking for them: a correction or counter-evidence. Replies are read before anything is shown; an approved reply shows your name, role, date, statement and link. Your email address is never shown and is deleted 30 days after the decision (privacy).

More factual claims by Eliezer Yudkowsky

  1. Mostly true Factual September 2025
    “And a blade of grass one notes is a solar powered, fully self-replicating general factory.”

    A blade of grass is a solar-powered, fully self-replicating general factory.

  2. Mostly true Factual 7 April 2023
    “In today’s world you can email DNA strings to laboratories that will produce proteins on demand, allowing an AI initially confined to the internet to build artificial life forms or bootstrap straight…”

    Today one can email DNA strings to laboratories that will produce proteins on demand.

  3. Mixed Factual 20 April 2026
    “I was observing that obvious-seeming hypothesis myself, sometime around the time of GPT 3.5.”

    Yudkowsky himself observed the persona-selection-like hypothesis around the time of GPT 3.5.

  4. Mostly true Factual 20 April 2026
    “The test-passing preference stands on its own once formed.”

    Once formed through RL training, an LLM's test-passing preference stands on its own and is not overridden by its inference of what the user wants.

  5. Mostly true Factual 20 April 2026
    “LLMs surely have neural circuits they learned to model text streams from fictional and nonfictional personas that are not lying, deceiving, cheating.”

    LLMs surely have neural circuits they learned to model text streams from fictional and nonfictional personas that are not lying, deceiving…

All of them

Verdicts are made automatically; how verdicts are made.