DOOMBENCH.COM
Safety and alignment

Altman warns superhuman machine intelligence could eliminate humanity

Sam Altman argued that unintended goals, resource acquisition, self-preservation, and recursive self-improvement could make superhuman machine intelligence an existential threat to humanity.

0 comments · 0 votesOpen discussion

Public discussion is readable by everyone. Sign in to comment, reply, or vote.

No comments yet. Start the discussion.

Send detailed or private feedback →
CURRENT ASSESSMENT · REVISION 1
TOWARD DOOM46confidence 60/100

Why it moved the index

A dated first-person mechanism-level argument links misgeneralized goals, instrumental behavior, and recursive improvement to possible human extinction. It is analysis rather than an observed incident.

Assessment history

  1. R1
    Toward 46 · confidence 60

    Adds a notable pre-2020 first-person statement that was absent from the durable evidence catalogue.

    14 Aug 2026

Share this page

DoomBench social sharing card for Altman warns superhuman machine intelligence could eliminate humanity.

  1. DoomBench assesses “Altman warns superhuman machine intelligence could eliminate humanity” as evidence moving toward doom, with magnitude 46 and confidence 60 out of 100 in the safety and alignment category.

  2. The DoomBench assessment of “Altman warns superhuman machine intelligence could eliminate humanity” is based on reporting from Sam Altman's blog and records the editorial rationale, source quality, attribution, and revision history.

  3. DoomBench summarizes “Altman warns superhuman machine intelligence could eliminate humanity” as follows: Sam Altman argued that unintended goals, resource acquisition, self-preservation, and recursive self-improvement could make...

    https://www.doombench.com/news/altman-warns-superhuman-machine-intelligence-could-eliminate-humanity-2015-02-25