ClaimsEliezer Yudkowsky

“Today there is still no shield that can deflect a nuclear explosion;”

Eliezer Yudkowsky, author of chapter 2008 Artificial Intelligence as a Positive and Negative Factor in Global Risk (MIRI)

Factual Other global risks

As of 2008 there is no shield that can deflect a nuclear explosion.

Mostly true True in substance, wrong in a detail.

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.

Said for an organisation they lead or work for; it counts on both records.

What would count as true or false

True: as of 2008, no deployed or demonstrated shield or armour could deflect or withstand a nuclear explosion's blast and effects at close range. Missile defence against delivery vehicles does not count as a shield against the explosion itself. Mostly true: no such shield existed, but hardened bunkers could survive at some distance, which the claim's wording glosses over. Mixed: not applicable unless sources disagree substantially. Mostly false: some technology existed in 2008 that arguably deflects nuclear explosion effects in meaningful cases. False: a working shield able to deflect a nuclear explosion existed by 2008.

Set before the research began.

Evidence (4)

  1. E1 Chapter XII BIOLOGICAL EFFECTS atomicarchive.com (hosting Glasstone & Dolan, The Effects of Nuclear Weapons) · 1 January 2022

    “These facts emphasize the influence of circumstances of exposure on the casualties produced by a nuclear weapon and indicate that shielding of some type can be an important factor in survival.”

  2. E2 Chapter XII BIOLOGICAL EFFECTS atomicarchive.com (hosting Glasstone & Dolan, The Effects of Nuclear Weapons) · 1 January 2022

    “This is particularly the case for heavier structures extending below as well as above ground; persons properly located in such buildings could be protected from blast and initial nuclear radiations as well as from thermal radiations.”

  3. E3 DESIGNING FOR BLAST PROTECTION U.S. Department of Energy Office of Scientific and Technical Information (record of a Nucleonics journal article)

    “Blast shields and channeling are shown to be two efficient methods of blast protection. Reflecting barriers, isolation air gaps, and energy-release duration are methods which are inadequate.”

  4. E4 Would a fallout shelter really protect you in a nuclear blast? Live Science · 9 February 2025

    “So, while a bunker only a few miles from an explosion wouldn't be very helpful, a good shelter dozens of miles from a blast could protect inhabitants from radiation for days.”

Reasoning

E4 (2025) says a bunker only a few miles from a nuclear explosion "wouldn't be very helpful", and a good shelter dozens of miles away could protect people from radiation. That suggests no shield could withstand a nuclear explosion at close range, then or later. Glasstone & Dolan (E1, E2) do say that shielding, including heavier structures that extend below ground, can protect people from blast and radiation. That fits the criteria's "mostly true" case: hardened shelters could protect people at some distance, which the claim's wording glosses over. E3 deals with blast-protection design, has no stated period and does not show a shield against a nuclear explosion.

Cite this claim

Eliezer Yudkowsky, “As of 2008 there is no shield that can deflect a nuclear explosion.” (said 2008). Verdict: mostly true. Reliability API, claim c_6f74b5f6ebf9, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_6f74b5f6ebf9

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 28 February 2018
    “even most scientists in the field had no idea how to build a nuclear bomb”

    Even most scientists in the field had no idea how to build a nuclear bomb until the Hiroshima headlines.

  2. 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.

  3. 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.

  4. 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…

  5. Mostly true Factual 20 April 2026
    “The predict-as-if-honest circuitry is contextual, invoked to predict some streams of text but not others.”

    LLMs' predict-as-if-honest circuitry is contextual, invoked for predicting some text streams but not others.

All of them

Verdicts are made automatically; how verdicts are made.