Safety and alignment

Eliezer Yudkowsky frames ASI alignment as an irretrievable one-shot problem

Yudkowsky argues that safe tests of weaker systems cannot validate alignment under the materially different conditions where an ASI could cause catastrophe, and a severe first failure would prevent a corrective retry.

0 comments · 0 votesOpen discussion

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

No comments yet. Start the discussion.

CURRENT ASSESSMENT · REVISION 1
TOWARD DOOM22confidence 56/100

Why it moved the index

The post adds a distinct technical mechanism: distribution shift separates survivable development tests from the first capability regime in which alignment failure could remove the possibility of recovery. It is a detailed first-person argument, but not an empirical demonstration that the predicted critical transition will occur.

AUDIT TRAIL

Assessment history

  1. R1
    Toward 22 · confidence 56

    Initial inclusion from a dated first-person source recovered during Eliezer Yudkowsky's historical backfill.

    13 Aug 2026