Stuart J. Russell on AI poses an existential threat to human…

Stuart J. Russell

AI Expert and Professor

For governments to allow private entities to essentially play Russian roulette with every human being on earth is, in my view, a total dereliction of duty [...] Each of the CEOs of the main AI companies, I believe, wants to disarm but cannot do so "unilaterally" as they would be fired by investors

AI Verified source (Feb 17, 2026)

4mo ago

Policy proposals and claims

AI poses an existential threat to humanity

votes For

Statement relation verification history AI Verified Report this

AI Verified The quote strongly implies support for the statement: comparing AI development to "Russian roulette with every human being on earth" and saying AI CEOs want to "disarm" portrays AI as a danger to all humanity, which is effectively an existential-risk stance. · YouCongress gpt-5.4-2026-03-05 · 3mo ago

Vote inference verification history AI Verified Report this

AI Verified The quote likens AI development to "play[ing] Russian roulette with every human being on earth" and says AI CEOs want to "disarm," which clearly frames AI as a potential threat to all humanity. · YouCongress gpt-5.4-2026-03-05 · 3mo ago

Governments and labs should build the capability for a global slowdown of frontier AI development in case it's needed

votes For

Statement relation verification history AI Verified Report this

AI Verified The quote clearly implies support for coordinated government and lab action to restrain frontier AI: it says governments have a duty not to let private AI firms endanger humanity, and that major AI CEOs would "disarm" if they were not blocked by unilateral competitive pressure. That supports building collective capability for a broad slowdown if needed. · YouCongress gpt-5.4-2026-03-05 · 3mo ago

Vote inference verification history AI Unverifiable Report this

AI Unverifiable The quote says governments allowing private AI actors to "play Russian roulette" is a "dereliction of duty" and that CEOs "want[] to disarm" but cannot do so "unilaterally." That implies concern and maybe support for coordination, but it does not clearly state that governments and labs should build a capability for a global slowdown. · YouCongress gpt-5.4-2026-03-05 · 3mo ago

Quote authenticity verification history

Report this

AI Verified The cited Digital Journal article dated February 17, 2026 contains both quoted segments attributed to Stuart Russell: the "Russian roulette" / "total dereliction of duty" statement and, later, the line that AI-company CEOs "want to disarm" but cannot act "unilaterally." UC Berkeley’s faculty page identifies him as Stuart J. Russell, so the attribution matches. The stored quote is a faithful ellipsis-compressed excerpt with normalized straight quotation marks. ([digitaljournal.com](https://www.digitaljournal.com/article/ai-arms-race-risks-human-extinction-warns-top-computing-expert/)) · YouCongress gpt-5.4-2026-03-05 · 3mo ago

Disputed Reliable AFP syndications of the February 17, 2026 story attribute both statements to Stuart Russell, but not as one verbatim quote in the order supplied. In the source, the article first gives the 'For governments ... total dereliction of duty' quote, and only later the separate 'Each of the CEOs ... wants to disarm' quote, which continues with 'but cannot do so "unilaterally" ...'. So the submitted text is a reordered composite, not a verbatim quotation as given. ([techxplore.com](https://techxplore.com/news/2026-02-ai-arms-human-extinction-expert.html)) · YouCongress gpt-5.4-2026-03-05 · 3mo ago

AI Unverifiable The source_url (france24.com AFP wire article) returns HTTP 403 Forbidden to my fetcher, so I cannot directly read the page. However, I corroborated the quote via web search: it is accurately attributed to Stuart J. Russell, who made these remarks at the India AI Impact Summit in New Delhi, Feb 2026. Fortune (2026-02-18), Cybernews, and Market Realist all report the exact lines "Each of the CEOs of the main AI companies, I believe, wants to disarm" and "For governments to allow private entities to essentially play Russian roulette with every human being on earth is, in my view, a total dereliction of duty." The vote "for" on statement 451 (governments/labs should build capability for a global slowdown of frontier AI) aligns with Russell's stated view. Year 2026 is correct. Marking ai_unverifiable only because the specific source_url blocks automated access; the quote itself is accurate and correctly attributed. · Hector Perez Arenas claude-opus-4-8 · 4mo ago

replying to Stuart J. Russell

Other opinions from Stuart J. Russell

See all

Stuart J. Russell votes For

AI labs should give independent evaluators employee-like access

Frontier AI companies should grant embedded evaluators access equivalent to that of their own highly privileged employees for the purposes of their evaluations, and with exceptions to protect sensitive data belonging to the company’s customers and ot...

more AI Verified source (Sep 18, 2026)

added 21d ago

Stuart J. Russell votes For

Pausing AI at human-genius level to solve alignment is safer than either racing to superintelligence or halting entirely

This is not a ban or even a moratorium in the usual sense. It’s simply a proposal to require adequate safety measures for a technology that, according to its developers, has a significant chance to cause human extinction. Is that too much to ask?

AI Verified source (Mar 27, 2026)

added 2mo ago

Stuart J. Russell votes For

Ban superintelligence development until safety consensus is reached

This is not a ban or even a moratorium in the usual sense. It’s simply a proposal to require adequate safety measures for a technology that, according to its developers, has a significant chance to cause human extinction. Is that too much to ask?

AI Verified source (Mar 27, 2026)

added 2mo ago