AI RESEARCH

Safety is Non-Compositional: A Formal Framework for Capability-Based AI Systems

arXiv CS.AI

ArXi:2603.15973v1 Announce Type: new This paper contains the first formal proof that safety is non-compositional in the presence of conjunctive capability dependencies: two agents each individually inca- pable of reaching any forbidden capability can, when combined, collectively reach a forbidden goal through an emergent conjunctive dependency.