Chile and Mexico sign pact for responsible AI development

The Chile–Mexico AI accord is a notable step toward a more multipolar AI landscape. By explicitly framing AI cooperation around “sovereignty” and shared regional infrastructure, both governments are signaling that they do not want Latin America’s AI future to be wholly intermediated by U.S. and Chinese hyperscalers. The commitment to co‑develop scientific talent, share compute resources like Mexico’s Coatlicue supercomputer, and align on responsible‑AI norms effectively seeds a regional AI bloc that can negotiate with big vendors from a position of greater strength. ([elespanol.com](https://www.elespanol.com/invertia/disruptores-innovadores/america-tech/mexico/20251227/chile-mexico-unen-fomentar-desarrollo-responsable-inteligencia-artificial/1003744054122_0.html))

For the race to AGI, this matters less because Chile or Mexico will suddenly field a frontier model and more because they are laying the institutional plumbing for local experimentation and governance. Access to serious compute, research collaborations and cross‑border datasets could support indigenous models tuned to Latin American languages, legal systems and social priorities. That diversity of approaches is healthy in a world where most AGI‑adjacent work is concentrated in a handful of Silicon Valley and Chinese firms. It also raises the bar for global players: if they want to deploy deeply in the region, they’ll likely have to plug into this emerging ecosystem rather than treating Latin America as a pure export market.