Latam-GPT: Latin America Builds Its Own Artificial Intelligence

Twelve Latin American countries are working together to launch Latam-GPT, a new artificial intelligence (AI) language model designed just for their region.

The project, led by Chile’s National Center for Artificial Intelligence (CENIA), brings together over 30 local institutions. They plan to release the model in September 2025.

Latam-GPT is different from most AI models because it focuses on Latin America’s unique cultures and languages. Most AI today is trained mainly on English and often does not understand local ways of speaking or cultural references.

Latam-GPT will use data from across the region, including Spanish, Portuguese, and even indigenous languages like Rapa Nui from Easter Island. The project started in January 2023.

It uses advanced technology called Llama 3 and is trained on computers based in Chile and other countries in the region. The regional development bank CAF and Amazon Web Services have supported the project, but there is no permanent budget yet.

Boosting AI for Latin America’s Unique Needs and Languages

Latam-GPT’s main goal is to help local businesses, governments, and schools use AI that understands their needs. For example, it can be used to build chatbots for public services or create educational tools that reflect local history and language.

One important part of the project is preserving native languages. The team has already built a translator for Rapa Nui and wants to add more indigenous languages in the future.

The model is open source, which means anyone can use and improve it. This gives Latin American countries more control over how they use AI, instead of relying on technology from other parts of the world.

The project uses more than 8 terabytes of local text to train the model, so it can give better answers about the region’s history, geography, and culture. By building Latam-GPT, Latin America hopes to become less dependent on foreign technology companies.

The project could help local industries, make public services more effective, and support innovation across the region. The future of Latam-GPT will depend on attracting more funding and showing that the technology works well for local needs.