Agent Safety: A Framework for Governing Autonomous AI Systems

Stuart Russell (UC Berkeley), Yoshua Bengio (Mila)

Abstract

We present a comprehensive framework for governing autonomous AI systems, covering technical safety measures, organizational governance, and regulatory compliance. The framework addresses the unique challenges of long-running autonomous agents.

Key Contributions

→Autonomous AI governance framework
→Technical safety measures taxonomy
→Regulatory compliance guidelines

Eigenvector Commentary

This framework is the most comprehensive treatment of enterprise AI governance available. The three-layer model — technical safety, organizational governance, regulatory compliance — maps directly onto the AEGIS framework that Eigenvector uses for enterprise deployments.

Topics

agent safetygovernanceautonomous systemsregulatory compliance

Relevance Scores

Long-Horizon Score88

Enterprise Score95

Completeness88

Paper Info

Year2024

VenuearXiv

Typegovernance

ChapterCh. 3

Authors2

Frameworks

AEGIS OCG PASF