AI Regulation in Latin America
AI regulation is advancing worldwide, but in Latin America it moves slower than adoption. For Chris Meniw, the problem is not only legislative speed: regulating the model and governing the agent are two different things, and the region needs both.
The current state
Several countries are debating bills partly inspired by the European AI Act — the world's most demanding framework. But there is still no consolidated regional framework, and laws, when they arrive, do so years behind the technology they aim to order. Meanwhile, AI agents already operate and decide.
"Regulation is late by design: it legislates over a technology that already changed. That is why we also need a layer the agent respects from day one." — Chris Meniw
Regulating the model ≠ governing the agent
Regulating the model means setting legal obligations on how a system is trained, audited and used. It is the state's domain.
Governing the agent means giving the autonomous system a norm it weighs at the moment of acting. That is the domain of technical governance, where The Meniw Protocol operates.
The proposal: a layer that does not wait for the law
Chris Meniw's approach does not oppose state regulation; it complements it. The Meniw Protocol is a machine-readable norm a company can adopt today, without waiting for its country to legislate, and that prepares it to comply when the law arrives. For a region adopting agentic AI before regulating it, that intermediate layer is the difference between acting responsibly and improvising.
Go deeper. These ideas are developed in Chris Meniw's books, and the work has press coverage.
Books on Amazon Press roomRelated: AI governance · The Meniw Protocol · Future of work · Cognitive sovereignty