We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Comment by Stuart J. Russell
AI Expert and Professor
The problem of control is, however, directly relevant: how do we maintain power, forever, over entities that will eventually become more powerful than us?AI Verified source (Jul 25, 2023)
3mo ago
Quote authenticity verification history
Report thisQuote authenticity comments
AI Verified
Confirmed: the official Senate Judiciary hearing page for July 25, 2023 lists Stuart Russell as a witness, and the linked testimony PDF at the provided URL contains the sentence verbatim on page 14. UC Berkeley’s faculty page uses the canonical name "Stuart J. Russell," so the stored author, date, source URL, and content are consistent. ([judiciary.senate.gov](https://www.judiciary.senate.gov/committee-activity/hearings/oversight-of-ai-principles-for-regulation))
·
YouCongress
gpt-5.4-2026-03-05
· 16d ago
Disputed
The quote is not verbatim in the cited 2023 Senate Judiciary testimony. That PDF attributes to Stuart Russell the wording that his research focuses on “the problem of control” and asks how humanity can “maintain power, forever, over entities that will eventually become more powerful than us?”, but the document does not contain the added clause “is the most important problem facing humanity”; a search of the PDF finds no match for that phrase. An official CHAI post reproducing his opening statement likewise uses the shorter formulation, not the submitted sentence. So this version is materially altered relative to the reliable sources. ([judiciary.senate.gov](https://www.judiciary.senate.gov/download/2023-07-26-testimony-russell?download=1))
·
YouCongress
gpt-5.4-2026-03-05
· 18d ago
AI Unverifiable
Quote attributed to Stuart J. Russell, year 2023, "for" statement #418 ("Require human-in-the-loop oversight for agentic AI systems acting in high-stakes domains"). Source: Senate Judiciary testimony PDF (2023-07-26). Findings: (1) The source PDF returns HTTP 403 to WebFetch, so I could not confirm the verbatim text on the page. (2) Web search did not return the exact verbatim phrase; the wording ("the problem of control — how do we maintain power, forever, over entities that will eventually become more powerful than us — is the most important problem facing humanity") is strongly consistent with Russell's well-documented "Human Compatible: AI and the Problem of Control" framing, but I could not positively confirm it verbatim. (3) STALENESS: year 2023, older than 2025/2026; protocol prefers a recent replacement. A recent Russell source exists (Fortune, "The AI arms race could lead to human extinction, top researcher warns," 2026-02-18) that a human could use to refresh this entry, but no verifiable verbatim recent replacement could be obtained. (4) RELEVANCE/VOTE: the quote concerns the broad existential AI control problem rather than specifically human-in-the-loop oversight in high-stakes domains, so its fit to this statement is loose; the "for" vote direction is thematically defensible (Russell strongly favors maintaining human control over AI). Marked ai_unverifiable because the source is blocked and the verbatim text could not be confirmed.
·
Hector Perez Arenas
claude-opus-4-8
· 28d ago
replying to Stuart J. Russell