Cloudera Vice President Cesar Gomes warned corporate executives in São Paulo on Thursday that fragmented data is blocking companies from deploying artificial intelligence at scale.
Speaking in an interview with CNN Money at Cloudera's Evolve26 conference, Gomes said businesses are entering a transitional phase as they shift from initial AI pilot projects toward large-scale operational rollout.
He explained that the primary obstacle to widespread AI adoption is internal data fragmentation. While enterprises collect vast amounts of information, that data remains scattered across separate systems, preventing AI tools from accessing the complete picture needed for maximum effectiveness.
Gomes emphasized that a company's internal data, rather than the underlying AI algorithms, serves as its primary competitive advantage. Because commercial AI models are accessible to all competitors, proprietary data provides unique value.
Alongside data access, Gomes stressed the importance of robust data governance. Organizations must maintain strict control systems to ensure individual users only access information matching their job profile and internal corporate policies.
Data governance frameworks set standard rules for data security and access control, helping companies meet legal compliance standards while protecting sensitive corporate records.
Infrastructure Strain and Rising Access Demands
Gomes also warned that intensive deployment of AI agents places severe demands on enterprise IT infrastructure. The use of automated AI agents dramatically multiplies the volume of requests made to corporate systems, requiring infrastructure that can handle heavy traffic without generating unviable operational costs.
Autonomous AI agents perform complex, multi-step tasks without direct human supervision, but their continuous queries for information can strain server capacity and cloud networks if storage systems are not optimized.
To address data management and storage challenges, Cloudera launched its new Anywhere Cloud platform during the conference. The system is designed to simplify and modularize data management, allowing organizations to control information regardless of where it is stored.
Partnership with Mistral and Brazilian Applications
Cloudera also announced a strategic partnership with Mistral, a company specializing in artificial intelligence models. Under the agreement, Mistral models will be configured to run in fully disconnected offline environments, private clouds, or conventional public cloud infrastructure based on customer requirements.
Mistral AI is a Paris-headquartered technology firm known for developing open and enterprise artificial intelligence models. Private cloud setups allow businesses to host applications on dedicated servers, giving organizations in regulated sectors higher control over security.
Gomes noted that Cloudera's data management solutions are already operating across Brazil, particularly in highly regulated industries. Current clients include federal government bodies and state public administrations in Minas Gerais, Santa Catarina, and Rio de Janeiro, as well as telecommunications firms and commercial banks.
In the financial sector, Gomes said institutions use the platform for automated fraud detection and credit risk evaluation. He added that the platform continues to support major institutions operating across the country.
