Chile Launches Latam-GPT: Latin America’s First Open-Source AI Model to Boost Regional Presence in Global AI Race

Chile Launches Latam-GPT to Strengthen Latin America’s Presence in AI

Chile on Tuesday inaugurated the first open-source artificial intelligence language model trained on Latin America's different cultures, with the goal of better reflecting regional reality and strengthening the region's position in the global AI competition. Latam-GPT is the product of a two-year regional effort spearheaded by Chile's National Center of Artificial Intelligence, CENIA, and backed by more than 30 institutions from eight Latin American countries.

"Artificial intelligence is the greatest technological revolution of recent times, and from Latin America and the Caribbean, it is strategic and urgent that we play a role," Chilean President Gabriel Boric said Tuesday after the launch, noting that the new system will be critical in adding Latin American data and identity to AI.

The initiative, which was announced at the Paris Artificial Intelligence Action Summit in February 2025, will begin in early 2023 to remove linguistic biases in models trained predominantly on English data. Rather than directly competing with consumer products like ChatGPT or Google Gemini, Latam-GPT serves as the platform for future regional applications. The program focuses on Chile's efforts to increase its influence in artificial intelligence while promoting regional collaboration to develop technology that reflects Latin America's language diversity and cultural realities.

Training Data and Cultural Representation in Latam-GPT

"Latam-GPT is trained with a proportion of Latin American data that previously did not exist online and was not included in existing models," stated Rodrigo Durán, executive director of CENIA. "This allows for more accurate, correct and efficient performance when it comes to Latin America and the Caribbean."

Latam-GPT is trained using data from private sources gathered through regional strategic partnerships, as well as synthetic data intended to target underrepresented areas, according to Gabriela Arriagada, a CENIA researcher and the project's ethics team leader. Latam-GPT required more than eight terabytes of data, which is comparable to millions of books.

"When we talk about incorporating Latin American culture, we are referring to a training approach designed to address data that reflects cultural realities, identifying where gaps exist in other models, understanding their shortcomings, and gradually building knowledge to improve that representation," says Arriagada.

For the time being, the initiative will be conducted mostly in Spanish and Portuguese, with hopes to include Indigenous languages in the future. According to the developers, this culturally targeted training strategy will assist ensure that the system more accurately reflects the region's languages, expertise, and social environment.

Regional Collaboration and Significance for Latin America

Latam-GPT is a "very important milestone for Latin America," according to Luis Chiruzzo, an engineering professor at the University of the Republic in Uruguay who is not engaged with the project. "That provides some assurance that, at the very least, everyone is included in the training," according to Chiruzzo.

According to Durán, executive director of CENIA, the creation of Latam-GPT has given the area the technical capacity to design AI models. "The fact that Latin America has banded together to form a working organization is an extremely encouraging indicator. It demonstrates that Latin America can develop and comprehend how to generate this technology, which has huge regulatory consequences since you cannot govern something you do not understand," he said.

Chiruzzo believes Latam-GPT will struggle to compete with larger firms that have more resources. "Still, it's an important step forward and will make it possible to start positioning ourselves in the world of language models with our own voice," according to Chiruzzo. The program emphasizes increased regional cooperation in technological development as well as the desire to play a more active role in the global AI ecosystem.

Global AI Competition and Chile’s Investment in Technology

The quest for AI leadership has prompted countries to reconsider their policies and strategies for AI development. According to Oxford University data, the United States, China, and the European Union account for more than half of the world's most powerful data centers for developing and supporting AI systems. The paper states that Africa and South America have essentially no AI clusters.

Chile has been rushing to extend its role in the AI boom in recent years, luring new talent and establishing new data centers. In June of last year, President Boric declared in his State of the Union speech that the country must embrace AI, adding that a "country that does not invest in artificial intelligence risks falling behind in tomorrow's global landscape."

Latam-GPT was created with just $550,000 from CENIA's budget and the Development Bank of Latin America (CAF). The company used Amazon Web Services' cloud to construct its first version, which will be available at the end of February. The succeeding versions will be trained on a supercomputer at the University of Tarapacá in northern Chile, which will cost around $4.5 million and will begin in the first semester of 2026.