Most Mexican Companies Do Not Know Where They Use AI: IBM

Eighty percent of Mexican organizations are accelerating AI adoption without maintaining visibility into how, where, and by whom these systems are deployed. Research from IBM reveals a governance gap that could increase operational, cybersecurity, compliance, and accountability risks as AI investments continue to grow across the country.

AI adoption is advancing rapidly across Mexico, but governance mechanisms are not keeping pace. According to IBM, only 20% of Mexican organizations know exactly which AI capabilities are deployed across their operations and where those systems operate within their technology infrastructure. As companies increase AI spending, limited visibility is emerging as a business risk.

The gap between adoption and governance is becoming more visible as organizations deploy AI across multiple business functions without maintaining centralized oversight. “Of the Mexican respondents in this AI governance study, only 20% reported that they know exactly which AI functionalities are being implemented and where they are located within their enterprise technology structure,” says Luis Felipe Guzmán, Data and AI leader, IBM Mexico.

The findings come from the AI in Motion research, conducted by the IBM Institute for Business Value and the Dubai Future Foundation. The study surveyed more than 1,000 senior executives across 21 industries and more than 20 countries.

Among Mexican respondents, only 13% maintain an inventory of their AI systems. Such inventories serve as a foundational governance mechanism, allowing organizations to identify which models are deployed, what data they process, what decisions they can influence, and which stakeholders are accountable when errors occur.

The results suggest that AI adoption is progressing faster than the controls designed to manage it. While organizations increasingly integrate Generative AI, intelligent assistants, and autonomous agents into daily operations, many lack a complete view of how these technologies interact with internal systems and sensitive information.

The challenge extends beyond officially approved applications. Employees can independently adopt publicly available AI services, connect them to business processes, or upload corporate information without authorization from information technology teams. This creates fragmented deployments that may operate outside established security controls. As a result, organizations struggle to reconstruct events following data leaks, compliance audits, operational failures, or disputes involving automated decisions.

The governance challenge is becoming more urgent as AI systems evolve from passive assistants into autonomous agents capable of executing actions, modifying records, and interacting directly with enterprise workflows. IBM's research indicates that only 3% of organizations operate a centralized platform capable of coordinating AI agents. Without these controls, enterprises face growing difficulties monitoring decisions, controlling costs, and managing risk exposure.

Investment Rises While Visibility Falls 

The governance gap emerges at a time when AI investment remains a strategic priority for Mexican businesses. According to the first edition of KPMG's Global AI Pulse survey, organizations in Mexico expect to invest an average of US$171 million in AI over the next 12 months. The study, published by KPMG International, found that 74% of Mexican organizations would continue prioritizing AI investments even under recessionary conditions.

At the same time, 72% of organizations in Mexico report that AI is already generating meaningful business value, exceeding the global average of 64%. However, the same research highlights challenges that closely align with IBM's findings. Companies report difficulties measuring AI-driven value, adapting governance models, managing data privacy, addressing cybersecurity risks, and overcoming workforce resistance to adoption.

“The true value of AI does not lie in how much organizations invest, but in how effectively they integrate it into their business operations,” says Gustavo Gómez, AI lead partner, KPMG Mexico.

The contrast is notable. Organizations are allocating substantial capital to AI initiatives while many still lack visibility into the systems already operating within their environments. The issue becomes increasingly significant as AI agents gain wider adoption. KPMG reports that 31% of Mexican organizations are already implementing and scaling AI agents, while 15% have reached a level of maturity where AI deployment generates measurable competitive advantages and business value.

Risk management is emerging as a defining factor separating high-performing organizations from the broader market. Globally, 75% of business leaders express concerns regarding data security, privacy, and AI-related risks.

AI is a Priority for Business and Security

Cybersecurity specialists warn that these concerns will intensify as AI capabilities expand. According to Kaspersky's forecasts for 2026, AI will reshape both cyber defense and cybercrime. The company expects deepfakes, autonomous attack tools, synthetic voice fraud, and AI-generated social engineering campaigns to become increasingly sophisticated and accessible.

These developments raise the stakes for organizations that lack visibility into their AI deployments. Kaspersky argues that governance can no longer be treated solely as a compliance requirement. Instead, organizations must integrate security and privacy-by-design principles throughout the AI lifecycle, from development and deployment to monitoring and retirement.

The company also recommends incorporating AI governance into enterprise risk frameworks, establishing clear internal policies, conducting impact assessments, implementing technical controls, and maintaining human oversight over critical business processes.

“Companies will operate in an environment where information can no longer be assumed authentic by default,” says Claudio Martinelli, General Manager for the Americas, Kaspersky. “The competitive advantage will increasingly depend on building resilient business models where AI usage aligns with risk management frameworks.”

For Mexican enterprises, the convergence of these findings points to a broader strategic challenge. AI adoption is generating measurable business outcomes and attracting significant investment. Yet visibility, governance, and accountability remain underdeveloped across many organizations.