Mexico has begun building public capabilities for artificial intelligence, but still lacks a comprehensive strategy that connects infrastructure, training, innovation, employment, and regulation. The Coatlicue project and recent legislative initiatives show concrete progress, though they also highlight the need for sustained national coordination.
The challenge is relevant because AI will transform business activities and public services, from customer service and education to health, security, and commerce. Without a common roadmap, the country may limit itself to importing solutions and reacting to isolated risks, instead of developing its own talent, technology, and applications.
Coatlicue represents one of the main infrastructure efforts. The Mexican government announced that the supercomputer will have a public investment of 6 billion pesos and that its construction will take place in 24 months starting from January 2026. The project seeks to expand national capacity to process data and conduct AI research.
However, computing capacity alone does not solve adoption gaps. Uneven fixed internet coverage, low digital literacy, lack of regional infrastructure, and cybersecurity risks can prevent universities, entrepreneurs, and micro, small and medium-sized enterprises from accessing these tools on equal terms.
The regulatory front is also advancing. Congress is examining mechanisms to monitor AI development and use, while proposals include digital literacy, algorithmic transparency, human oversight, and supervisory criteria. The Presidency, for its part, announced regional forums to build a proposal on digital platforms, social networks, and artificial intelligence.
Regulation must protect rights without turning technology into the object of a purely punitive response. For companies, this means adopting governance models, defining permissions, documenting processes, and measuring results before scaling automation. It also requires distinguishing between an AI tool and the responsibilities of those who design, implement, or use it.
In the private sector, readiness can begin with verifiable use cases: automating repetitive tasks, centralizing conversations, analyzing business data, and improving customer experience. At Onix Board we integrate multichannel conversational automation, unified inbox, e-commerce management, AI-assisted content generation, and analytics dashboards to turn these capabilities into measurable operational flows.
Responsible adoption does not depend on incorporating as many tools as possible, but on choosing concrete processes, establishing indicators, and maintaining human controls. Organizations can start with bounded pilots and assess response times, data quality, error reduction, team adoption, and business results before expanding implementation.
Mexico already has important pieces: a federal digital agency, a supercomputing project, legislative activity, and an expanding business ecosystem. The next step is to integrate them into a State policy that trains talent, facilitates innovation, reduces regional gaps, and provides certainty for the responsible use of AI. As that strategy consolidates, companies can move forward with practical solutions, clear metrics, and controls that enable AI to become real productivity.