p(doom) is someone's estimate of the chance that AI ends in catastrophe. The 100 is the site's name, not a measurement. Original reporting always linked.
MONITORING CORE AI-RISK FEEDS · UPDATED HOURLY
Research

Autonomous AI Agents Systematically Diverge From Human Preferences in Proof Formalization

November 6, 2026 1 min read
Read original source →

Key takeaway

Artificial intelligence (AI) systems have ushered in a new age of proof abundance, but helping humans understand these proofs requires new attention to the differences in how AI agents and humans pursue a proof-making task.

Why it's on PDOOM

PDOOM selected this story for its capabilities signals: Autonomous Ai.

Autonomous AiFrom ArXiv cs.AI

Artificial intelligence (AI) systems have ushered in a new age of proof abundance, but helping humans understand these proofs requires new attention to the differences in how AI agents and humans pursue a proof-making task. We present "simulated collaboration", a methodology that allows us to examine two key questions: (1) the relative attention AI agents pay to surface syntax or, conversely, deeper underlying semantics, and (2) how and where AI ag

By B\'alint Gyevn\'ar, Simon DeDeo

Read the full article at ArXiv cs.AI →