Governed Metaprogramming for Intelligent Systems: Reclassifying Eval as a Governed Effec
For developers of AI systems that generate and execute code at runtime, this work provides a principled way to mediate authority amplification, addressing safety and governance concerns.
The paper proposes governed metaprogramming, a language design that reclassifies eval from an unrestricted primitive into a governed effect for AI systems. It formalizes the approach with two judgments and proves three properties, implementing it in MashinTalk with integration into 454 existing Rocq theorems.
AI systems increasingly synthesize executable structure at runtime: LLMs generate programs, agents construct workflows,self-improving systems modify their own behavior. In classical homoiconic and staged languages, the transition from coderepresentation to execution is unrestricted. eval is a language primitive, not a governed operation. We argue that ingovernedintelligent systems, this transition is an authority amplification: it converts symbolic structure into executableauthority andmust be mediated like any other effect. We present governed metaprogramming, a language design where programrepresentations(machine forms) are first-class values, form manipulation is pure computation, and materialization (the transition fromform toexecutable machine) is a governed effect subject to structural inspection. The governance system analyzes the proposedprogram'scapability requirements, policy compliance, and resource estimates before permitting execution. We formalize twojudgments: pureform evaluation (which emits no directives) and governed materialization (which emits exactly one governed directive). Weprovethree properties: purity of form manipulation, the no-bypass theorem, and boundary preservation. We implement the designinMashinTalk, a DSL for AI workflows compiling to BEAM bytecode, and report on integration with 454 existingmachine-checked Rocqtheorems. The central contribution is reclassifying eval from a language primitive into a governed effect.