The Agent Use of Agent Beings: Agent Cybernetics Is the Missing Science of Foundation Agents
For researchers and engineers developing LLM-based agents, this work offers a theoretical scaffold to move beyond empirical trial-and-error, but it is primarily a conceptual proposal without empirical validation.
The paper argues that LLM-based foundation agents lack a theoretical foundation and proposes Agent Cybernetics, a framework derived from classical cybernetics, to provide principles for reliability, lifelong running, and self-improvement. It demonstrates the framework's utility by analyzing failure modes in code generation, computer use, and automated research.
LLM-based foundation agents that perceive, reason, and act across thousands of reasoning steps are rapidly becoming the dominant paradigm for deploying artificial intelligence in open-ended, long-horizon complex tasks. Despite this significance, the field remains overwhelmingly engineering-driven. Engineering practice has converged on useful primitives (tool loops, memory banks, harnesses, reflection steps), yet these are assembled by empirical trial and error rather than from first principles. Fundamental questions remain open: under what conditions does a long-running agent remain on-task? How should an agent respond when its environment exceeds its representational capacity? What architectural properties are necessary for safe self-improvement? We argue that cybernetics, the mid-twentieth-century science of control and communication in complex systems, provides the missing theoretical scaffold for foundation agents. By mapping six canonical laws of classical cybernetics onto six agent design principles, and synthesizing those principles into three engineering desiderata (reliability, lifelong running, and self-Improvement), we arrive at a framework termed Agent Cybernetics. Three application domains, code generation, computer use and automated research, exemplify the analytical framework of agent cybernetics by identifying failure modes and concrete engineering recommendations. We hope that agent cybernetics opens a new research venue and establishes the scientific foundation that foundation agents need for principled, reliable real-world deployment.