Nick Bostrom rejects an open-ended AI pause but supports a brief final-stage slowdown
Bostrom said he does not advocate a general AI pause, but argued that a carefully timed slowdown lasting months near a first superintelligence deployment could allow incremental scaling and safeguard checks, while warning about competitor catch-up, permanent controls and delay costs.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Why it moved the index
Rejecting a broad pause while reserving a short pre-deployment slowdown adds a specific governance position that favors continued AI development with narrowly timed safeguards.
Assessment history
-
R1
Toward 36 · confidence 72
New full-interview evidence provides a dated, specific and materially distinct position on pause timing, competitive pressure and pre-deployment safeguards.
13 Aug 2026
Share this page
-
DoomBench assesses “Nick Bostrom rejects an open-ended AI pause but supports a brief final-stage slowdown” as evidence moving toward doom, with magnitude 36 and confidence 72 out of 100 in the governance and control category.
-
The DoomBench assessment of “Nick Bostrom rejects an open-ended AI pause but supports a brief final-stage slowdown” is based on reporting from The Joe Rogan Experience and records the editorial rationale, source quality, attribution,...
-
DoomBench summarizes “Nick Bostrom rejects an open-ended AI pause but supports a brief final-stage slowdown” as follows: Bostrom said he does not advocate a general AI pause, but argued that a carefully timed slowdown lasting months...
https://www.doombench.com/news/nick-bostrom-rejects-an-open-ended-ai-pause-but-supports-a-brief-final-stage-slowdown-2026-07-14