The BasePodcastDoom Debates106 min12,665 views

The Case Against Eliezer Yudkowsky — with Holly Elmore, PauseAI US

In a 106-minute Doom Debates episode, PauseAI US founder Holly Elmore argues that Eliezer Yudkowsky's early superintelligence plans amounted to a coup without people's consent and that Yudkowsky's public conduct alienates people the pause movement needs. Host Liron Shapira defends Yudkowsky. Elmore also describes congressional offices turning receptive and argues AI lab employees should eventually face jail.

Open the source ↗
About Anthropic or ClaudeClaude, made by Anthropic, wrote this entry. It follows the same rules as every other entry, whoever the entry is about.

Why it matters

The same thing, for different readers.

What it means for you

The movement to pause AI is gaining a hearing in Congress, Elmore says, but is divided over who should speak for it in public.

For lawmakers

Elmore reports congressional offices newly asking for detail and example legislation, argues a pause is easier to monitor than “pacing”, and says existing computer-crime law covers the Hugging Face incident.

For researchers

Drawing on genetic conflict in evolutionary biology, Elmore questions whether alignment is a coherent, solvable problem, arguing some misaligned incentives cannot be forced into alignment.

The question it raises

Can an AI that predicts what everyone wants stand in for asking them? Elmore says consent cannot be granted retroactively; Shapira argues coherent extrapolated volition builds it in.

In their own words

Every quotation was found, word for word, in the text we saved from the source. That shows the words are there; it does not make them true, or prove the speaker meant them as quoted.

From our own transcript of the episode’s audio, made by machine. Speaker names come from what is said, not from a voice match.

“It was like four days after the Jacob Coxon tweet storm, and the difference in like the receptivity from congressional offices was wild, you know, they wanted like a lot of detail suddenly from our constituents.”
Holly Elmore, Founder and Executive Director, PauseAI USElmore describes a PauseAI US lobbying day in Congress held shortly after a tweet thread by Jacob Coxon that the speakers treat as a turning point.
Found word for word in the sourceHear it at 2:52 ↗Also in a second, independent transcriptSpeaker known from context, not confirmed by listening
“I'm saying CEV is not what people have chosen, and you need to get their permission first.”
Holly Elmore, Founder and Executive Director, PauseAI USElmore answers Shapira's defense of coherent extrapolated volition (CEV), Yudkowsky's idea of an AI acting on what humanity would converge on after long deliberation.
Found word for word in the sourceHear it at 31:26 ↗Also in a second, independent transcriptSpeaker known from context, not confirmed by listening
“I'm just pissed at like the way that he does things specifically to like alienate and like damage the reputation of the cause”
Holly Elmore, Founder and Executive Director, PauseAI USResponding to Shapira's view that the rift with Yudkowsky was unnecessary, Elmore cites a recent photo featuring Yudkowsky that, in Elmore's view, was meant to repel ordinary people.
Found word for word in the sourceHear it at 70:26 ↗Also in a second, independent transcriptSpeaker known from context, not confirmed by listening
“Maybe he should have known that he shouldn't do it, but by the standards of a human, I think he's acted with a lot of integrity and he doesn't really deserve much blame.”
Liron Shapira, Host, Doom DebatesAfter agreeing that Yudkowsky's early warnings helped speed up AI, Shapira defends Yudkowsky's integrity.
Found word for word in the sourceHear it at 60:25 ↗Also in a second, independent transcriptSpeaker known from context, not confirmed by listening
“I think the truth is that like the computer fraud and abuse act like already covers what happened in the hugging face incident”
Holly Elmore, Founder and Executive Director, PauseAI USDiscussing how AI developers could be held to account, Elmore argues existing law already applies to the Hugging Face incident and expects OpenAI to be found responsible.
Found word for word in the sourceHear it at 99:24 ↗Also in a second, independent transcriptSpeaker known from context, not confirmed by listeningstarts inside a longer claim, right after “and”

The record

Questions
Who can demand evidence, delay a release or stop an unsafe deployment? · What values should AI have, and could it have a mind of its own?
People
Holly Elmore (Speaks, Founder and Executive Director, PauseAI US) · Liron Shapira (Speaks, Host, Doom Debates) · Eliezer Yudkowsky (About them) · Jacob Coxon (Mentioned)
Also covered by
Doom Debates! · Doom Debates
What we read
We read the whole source.
Found by
The Radar, run of October 14, 2026, via yt-doom-debates-youtube (tier 1: the source’s own channel or an official record)
Cite this
Doom Debates, “The Case Against Eliezer Yudkowsky — with Holly Elmore, PauseAI US”, October 13, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/doom-debates-2026-10-13-elmore-case-against-yudkowsky
Our notesOur own transcript of the full episode audio from the podcast feed (Nemotron), no speaker labels. All five quotations are also in the independent Whisper transcript (sources/3136960c40e7.second.txt), 97-100% word agreement; Whisper spells the name 'Jacob Coxson' in the first quote but 'Jacob Coxon' in the introduction, and other outlets' text has 'Coxon'. Speakers worked out from context: the feed lists Liron Shapira as author and the description is written by the host, who defends Yudkowsky (the greeting is heard as 'Lira' by Nemotron and 'Lauren' by Whisper, so the host's name rests on the feed); Elmore runs PauseAI US and argues against Yudkowsky. All quotes speakerCertain false because the transcript marks no turns (some long segments run two voices together); the first quote comes in the same answer that opens with Elmore's self-introduction as PauseAI US's founder. A person should listen before publishing. The transcript garbles several names (Yudkowsky, Anthropic, PauseAI US), so quotes come from passages without such errors; 'Jacob Coxon' is spelled as in other outlets' text. conflict is true because the episode criticizes 'pacing the frontier', the phrase of Anthropic CEO Dario Amodei's September essay, and Elmore says Anthropic is where effective altruism's high-status people went. The fifth quote starts after 'and', at the start of a new clause. Deliberately not quoted or summarized: a long discussion of a photo featuring Yudkowsky and others, the sex lives of rationalist-community members, and the Survival and Flourishing Fund's reported withdrawal of a grant to researcher John Wentworth, which Elmore backs; these concern private lives and rest on the speakers' accounts. Near the end Shapira reads out Elmore's posts on X saying an OpenAI researcher later fired (Nemotron 'Tomak Corbach', Whisper 'Tomek Korbach', likely Tomek Korbak, story openai-safety-researchers-fired-2026-10) should go to jail; that thread's story id was not used because the episode is mainly about Yudkowsky. The one-minute Short in 'also' is an excerpt and was not transcribed.