Colombia publishes ethical AI guide to govern public-sector deployments

Colombia’s new Ethical AI Guide is a notable example of a middle‑income democracy trying to front‑run AI governance rather than playing catch‑up. The document doesn’t just gesture at principles; it explicitly tells ministries and local agencies how to design, procure and operate AI systems so that human decision‑makers stay in the loop, algorithmic bias is monitored, and privacy and transparency are treated as hard requirements, not afterthoughts. It aligns local practice with OECD, UNESCO and UN guidance while adapting to Colombia’s own legal and institutional context.([dnp.gov.co](https://www.dnp.gov.co/Prensa_/Noticias/Paginas/gobierno-impulsa-guia-etica-de-inteligencia-artificial.aspx))

In the race to AGI, this kind of groundwork matters more than it might seem. As frontier models become commoditized, the real differentiation shifts to how states integrate them into critical functions—welfare systems, policing, tax, healthcare—without undermining legitimacy. Countries that invest early in governance capacity will be better positioned to harness powerful models safely when they arrive, instead of scrambling after a scandal.

It’s also a signal to Latin America and the Global South: you don’t need to wait for Brussels or Washington to define the rules of the game. A thoughtful, locally grounded guide from Bogotá can be a template for neighbors facing similar capacity and rights‑protection challenges.