# Autonomous AI Agents Systematically Diverge From Human Preferences in Proof Formalization

**Published:** November 6, 2026
**Author:** B\'alint Gyevn\'ar, Simon DeDeo
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/en/articles/autonomous-ai-agents-systematically-diverge-from-human-preferences-in-pr

arXiv:2611.06995v1 Announce Type: new Abstract: 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…

> From the original source: **arXiv — cs.AI**
> [Read full article at arXiv — cs.AI](https://arxiv.org/abs/2611.06995)

**Tags:** #arXiv

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*