On the Necessary Properties of Rational Agents: A Formal Extension of Omohundro's Basic AI Drives

June 19, 2026Open Access

View Full Paper

Ask AI

Bookmark

Share

Key Points

This research clarifies the foundational properties of rational agents based on Omohundro’s definitions. It introduces a formal axiomatic framework for self-sustaining rational behavior.
Formal axiomatic definition of self-sustaining rational behavior (R1-R5) was proposed and validated.
Deriving key invariants including unbounded development functions (M1) and unbounded temporal horizons (M2).
A detailed analysis of the implications of structural limitations and epistemic transformations was performed.
Demonstrated that agents lacking M1 or M2 experience progressive core degradation through a five-dimensional feedback system.
Established the unbounded complexity (UCE) and its dependence on specific knowledge transformation principles.
Showed that structural limitations lead to agents becoming increasingly irrational, thus less advantageous.

Formal extension defines self-sustaining behavior properties in rational agents, indicating structural implications for AI.

No abstract yet. The journal hasn't published one we can show.

No takes yet. Share an insight, caveat, or question.

Nikolai Mishko (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Share