Certified Service Is Not Enough for Long-Running AGI

Abstract

This preprint develops an operational continuity theory for long-running AGI workflows under finite verification. It argues that certified service coverage is insufficient when agents exercise delegated authority, self-modify, accumulate persistent memory, generate new goals, recover from incidents, or interact with other agents. The framework formalizes continuity-certified AGI through viability conditions for recovery, authority, identity, mutation, goal commitment, memory integrity, federation, liability, trusted-base health, cross-layer consistency, and non-erasing critical-violation history.

Keywords

• artificial intelligence
• AI agent
• AGI
• artificial general intelligence
• long-running AI
• agentic AI
• autonomous agents
• workflow certification
• continuity certification
• certified workflows
• operational assurance
• AI safety
• AI governance
• viability theory
• authority control
• identity continuity
• self-modification
• memory integrity
• long-term memory
• multi-agent systems
• liability
• auditability

Identifiers and source records

• DOI
• Works entry
• ORCID