Although Mexico has 50 data center facilities, none currently have the specialized infrastructure required to host large-scale artificial intelligence (AI) models. As a result, the computing capacity used by the country for this technology remains located mainly in the United States, China, and the Netherlands, said Adriana Rivera Cerecedo, executive director of the Mexican Data Centers Association (MEXDC).
During the presentation of studies on AI adoption in three strategic sectors for the country’s development, held at the Faculty of Engineering of the National Autonomous University of Mexico (UNAM), Rivera Cerecedo explained that without sufficient electrical capacity and telecommunications infrastructure, it will be impossible to develop a national artificial intelligence industry.
She explained that the infrastructure where AI models are hosted largely determines processing capacity, response speed, and the possibility of developing national applications. Therefore, having this capacity within Mexican territory is essential.
She identified two major challenges: strengthening the electricity supply and improving connectivity.
“Today, more than 552 municipalities in the country are still not connected, and those that are connected do not have broadband access,” she said.
However, she acknowledged that Mexico has made important advances and mentioned the interest of Mexican and foreign companies in deploying aerial and underground fiber networks, as well as the government’s efforts to expand energy generation and distribution.
“We also have millions of dollars invested in the Federal Electricity Commission because it benefits us to have modern electrical infrastructure. What we do alone is not enough, which is why the partnership between government, academia, and the private sector is important.”
She stated that building a data center in Mexico costs four times more than in Brazil, Mexico’s main competitor in Latin America.
“If a data center costs 100 million dollars, here it can reach up to 400 million due to higher infrastructure costs for electricity and connectivity,” she explained.
Infrastructure and Human Talent
Infrastructure is not the only challenge; Mexico must also encourage the development of specialized talent.
“We have a projection of 113,000 specialized and well-paid jobs between 2020 and 2031, but our young people want to become TikTokers, YouTubers, and influencers, but they are not getting involved in operations,” Rivera Cerecedo said.
The MEXDC has agreements with 10 universities across the country. However, when students visit data centers to understand their importance, “they are more concerned about taking a selfie, and that worries us a lot because we are going through a generational transition. The jobs linked to this sector, which are currently limited, will grow exponentially, and we need to prepare them.”
Querétaro continues to be a major destination for installing data centers; however, the MEXDC seeks to diversify the market.
“We want to look toward the states of Hidalgo, the State of Mexico, Nuevo León, Jalisco, and Yucatán, which could become important development hubs.”
Regarding water consumption, Rivera Cerecedo said that comparative studies conducted in Mexico show that the water footprint of using one megawatt in a data center for one year is equivalent to the amount of water needed to produce 2.4 kilograms of beef or 6.6 kilograms of pork.
“Twenty years ago, these centers were cooled with water, but science has helped us overcome that issue.”
She explained that the main concern is not water but solving the energy challenge, since fossil fuels have not yet been fully replaced.
Regarding criticism of data centers, especially in the United States where more than 800 organizations oppose them, she called for greater public awareness.
“These are infrastructures that will not stop. There has not been such a different change in the way we live since the Industrial Revolution 200 years ago. Therefore, we need to open the discussion about the sustainability of data centers.”
Rivera Cerecedo reported that three global companies are developing projects to host AI infrastructure in Mexico, with estimated investments of 82 billion dollars between 2020 and 2031.
“The ideal situation is for AI developed in Mexico to be established here. We are working toward that,” she concluded.
AI Adoption as a Strategic Tool to Increase Productivity
“Artificial intelligence is not a futuristic promise, but rather the circulatory system of the digital economy. That is why it is essential to stop guessing and start measuring,” said Salma Jalife Villalón, president of Centro México Digital, during the presentation of studies on AI adoption in three strategic sectors for the country.
“AI adoption is the strategic tool to multiply productivity and create value in the Mexican economy. It should not be understood as an isolated technological goal, but as a political, economic, and management decision,” she stated.
She considered that whether this technology becomes an engine for inclusion and widespread productivity or a source of greater inequality depends on today’s decisions regarding regulation, talent, and data governance.
Hugo Martínez, director of Google Mexico, said that studies on AI adoption in three strategic sectors —financial services and insurance; media and telecommunications; and electricity, water, and natural gas distribution— will allow for a broader and better-informed discussion.
Public Policies and AI Adoption
“It will provide us with a better understanding of how to design public policies that allow us to take better advantage of artificial intelligence. In a country like Mexico, which has significant gaps, the use of AI can have a positive impact,” Martínez explained.
In the financial services and insurance sector, which contributes 4.1% of Mexico’s GDP and generates 890,000 jobs, one of the most relevant findings shows that average AI adoption is 8.4%.
However, there is a significant gap between institutions:
- Commercial banks lead AI adoption with 31.1%.
- Insurance companies follow with 27%.
- Popular credit institutions reach 12.8%.
- Retirement savings funds only reach 3.9%.
The main challenges for expanding AI adoption in this sector include the lack of a sector-wide strategy, the absence of secondary open finance regulations for transactional data, and the shortage of specialized talent.
Media and Telecommunications Sector
The media and telecommunications sector, which generates 472 billion pesos annually, is the fastest adopter of AI, with 18% of companies already using the technology.
This figure is more than double the national average of 8%.
However, despite its leadership, Mexico remains 38.6 percentage points below the OECD average, which stands at 57.3%, and is equal to or below regional competitors such as Brazil, where AI adoption reaches 38.2%.
The challenges in this sector are related to regulatory conditions, infrastructure, and talent.
The value of AI is limited by outdated telecommunications systems and valuable editorial, biographical, and audiovisual archives that have not yet been digitized or properly classified.
Electricity, Water, and Gas Distribution
In the electricity, water, and natural gas distribution sector, the study highlights that the relationship with technology works both ways.
AI requires energy and water to operate, but at the same time, its adoption can make these networks more efficient, reliable, and less expensive.
In this sector, AI adoption reaches 5.7% and does not follow traditional market logic, but rather institutional structures shaped by regulation:
- Public-private participation in electricity.
- Gas price regulation.
- Fragmentation of more than 2,800 municipal public water organizations.
The electricity subsector has a higher AI adoption rate, reaching 25.2%, while the water sector remains low at 5.7%.
For Centro México Digital, Mexico’s real challenge is no longer understanding the potential of artificial intelligence, but creating the regulatory, infrastructure, and talent conditions necessary to accelerate adoption and prevent the country from widening the gap with other economies.
“Analyzing these sectors is not merely an academic exercise; it is the thermometer that shows our competitiveness in the world. Knowing where we stand allows us to precisely determine where we need to go,” Jalife Villalón concluded.
Source: jornada
