Colombia Ranks Fourth in Latin American Artificial Intelligence Index
In the latest Latin American Artificial Intelligence Index (ILIA), Colombia ranks fourth, after Chile, Brazil and Uruguay. It ranks well above Mexico, which is in eighth place.
The ILIA measures how ready 20 countries in Latin America and the Caribbean are to use artificial intelligence (AI), how much they have developed it, and how they manage and regulate it. The index is promoted by Chile’s National Center for Artificial Intelligence (CENIA) and the Economic Commission for Latin America and the Caribbean (ECLAC), with support from other public and private organizations.
Three groups: pioneers, adopters and explorers
The ranking has a maximum score of 100 points. It divides countries into three groups based on how far they have progressed: pioneers, adopters and explorers.
The pioneers score more than 60 points. They are Chile (71.4), Brazil (68.4) and Uruguay (67.0). These countries show “a high and relatively consolidated performance in infrastructure, talent, research, innovation and governance.”
The adopters, led by Colombia, have developed important skills and resources but still have gaps that “limit a broader deployment of AI.” This group includes countries scoring between 35 and 60 points: Colombia (56.8), Costa Rica (56.0), Argentina (55.6), Peru (52.7), Mexico (52.0), Ecuador (51.8), the Dominican Republic (45.1), Panama (39.3), El Salvador (37.9) and Trinidad and Tobago (35.5).
The third group, called explorers, includes countries that are at an early stage of development and face limits in infrastructure, research, AI use or the way AI is managed and regulated. Countries with scores below 35 points belong to this group: Paraguay (32.6), Jamaica (32.4), Cuba (30.9), Guatemala (29.9), Bolivia (26.4), Honduras (26.2) and Venezuela (23.2).
In its 2026 ILIA report, ECLAC said that mobile network coverage and access to 5G are important for using AI applications. The countries with the widest coverage are Uruguay, the Dominican Republic, Brazil, Chile, Guatemala, Mexico, Colombia and Peru.
The report also looks at professional training for AI experts. It says that “the region has an average of 12.4 points out of a total of 100 points. The best-positioned countries are Paraguay (28.2 points), Brazil (24.5), Costa Rica (22.6), Colombia (19.6), Mexico (15.8), Trinidad and Tobago (14.6) and Chile (14), all above the regional average.”
Regional cooperation is important for the future
The ILIA warns that having control over AI does not mean a country must be able to build every technology on its own. Instead, it means countries can make their own decisions and keep different options available in a world where technologies depend heavily on one another.
For Latin America and the Caribbean, this means improving local skills and resources while building relationships with other countries. These relationships should help countries keep their options open, have more freedom to act and strengthen their ability to take action.
In this situation, working together across the region becomes strategically important. Agreements and organizations that help countries share infrastructure, knowledge, skilled people and technology can help solve problems that would be difficult for one country to handle alone. They can also give the region more influence when dealing with global players.
From this point of view, control over AI is measured not only by how independent a country is from others. It also depends on how well the region can act in a system where countries rely on one another.
For the ILIA, this means that measuring the skills and resources countries already have is not enough. It is also important to understand their technology relationships. These relationships affect how much freedom countries have to form partnerships, keep and support local alternatives, and stand by their decisions.
Connectivity versus computing power
The report says the gap between Latin America and the Caribbean and countries outside the region is smaller for connectivity than for other advanced digital infrastructure. On some measures, countries in the region even perform better than countries outside it.
However, the gap becomes much larger when looking at the computing power available to develop and use AI systems. Even when the comparison includes only the nine countries in the region that have installed computing capacity, the average for countries outside the region is between 11 and 51 times higher, depending on the measure. When all 20 countries in the region are included, the difference rises to between 24 and 113 times.
The problem is made worse because computing resources are concentrated in a few countries. Eleven countries have no infrastructure in any of the four areas studied. Brazil has 92% of the region’s GPU capacity, and Brazil, Ecuador and Chile together account for 97.4%.
The comparison shows two different situations. Latin America and the Caribbean is getting closer to countries outside the region in connectivity, but the gap in computing power is much larger. The computing infrastructure that the region does have is also concentrated in only a few countries.