Data study ·
Red Hat warns of a dangerous gap between confidence and actual control of AI in Spanish companies
Original title, in Spanish: “Red Hat alerta de una peligrosa brecha entre la confianza y el control real de la IA en las empresas españolas”
- Publication
- IT User (Spain)
- Author
- Unsigned
- Published on
- 27 October 2026 · read in the page metadata · updated on 27 October 2026
Summary
A Red Hat study, presented by Julia Bernal, shows that 100% of the Spanish organisations surveyed have already deployed AI agents in production. However, the study reveals a gap between the confidence expressed by CIOs (85% say they are confident in the security of their agents) and reality: only 30% know precisely how their data flows, 22% have a complete view of their software supply chain, and 28% continuously monitor supplier risks after deployment. The study identifies four pillars required for governance (data traceability, software traceability, ecosystem transparency, agent lifecycle control) and warns of the risk of uncontrolled proliferation of agents without a clear identity or oversight. Red Hat links these issues to digital sovereignty, which is now discussed at board level.
Summary based on the full text of the article.
Read the article on IT User →Figures cited
-
100 %
share of Spanish organisations that have deployed AI agents in a production environment
Spanish organisations surveyed, Red Hat 2026 study
« el 100% de las organizaciones españolas consultadas ya ha desplegado agentes de inteligencia artificial en entornos productivos. »
Source: Étude Red Hat
-
85 %
share of Spanish CIOs confident in the security and compliance of their AI agents
Spanish CIOs, Red Hat 2026 study
« El 85% de los CIO españoles asegura confiar en que sus agentes de IA funcionan de forma segura y conforme a las normativas aplicables. »
Source: Étude Red Hat
-
30 %
share of organisations that know precisely how data flows and where it is stored
Spanish organisations, Red Hat 2026 study
« únicamente el 30% afirma conocer con precisión los flujos de datos dentro de su organización y los lugares donde estos se almacenan y procesan. »
Source: Étude Red Hat
-
22 %
share of organisations with a comprehensive view of software supply chain dependencies
Spanish organisations, Red Hat 2026 study
« solo el 22% dispone de una visión exhaustiva de las dependencias directas e indirectas de su cadena de suministro de software »
Source: Étude Red Hat
-
28 %
share of organisations continuously monitoring AI supplier risks after deployment
Spanish organisations, Red Hat 2026 study
« apenas el 28% monitoriza continuamente los riesgos asociados a proveedores de IA después de su implantación. »
Source: Étude Red Hat
-
47 %
share of organisations acknowledging blind spots in data traceability
Spanish organisations, Red Hat 2026 study
« El 47% reconoce que existen puntos ciegos en la trazabilidad de los datos »
Source: Étude Red Hat
-
22 %
share of organisations that have integrated AI capabilities without fully understanding data flows
Spanish organisations, Red Hat 2026 study
« el 22% admite haber incorporado capacidades de IA sin comprender completamente cómo fluyen los datos entre sistemas y modelos. »
Source: Étude Red Hat
-
79 %
share of Spanish companies able to apply critical patches within 24 hours
Spanish companies, Red Hat 2026 study
« Aunque el 79% de las empresas españolas afirma ser capaz de aplicar parches críticos en menos de 24 horas después de su publicación »
Source: Étude Red Hat
-
22 %
share of organisations with an up-to-date view of all software components, versions, origins and dependencies
Spanish organisations, Red Hat 2026 study
« Solo el 22% dispone de una visión actualizada de todos los componentes de software, sus versiones, orígenes y dependencias. »
Source: Étude Red Hat
-
28 %
share of organisations applying measurable criteria and continuous monitoring to AI suppliers
Spanish organisations, Red Hat 2026 study
« Solo el 28% aplica criterios medibles y monitorización continua sobre sus proveedores de IA una vez implantadas las soluciones »
Source: Étude Red Hat
-
34 %
share of organisations carrying out thorough assessments at the start of a project but struggling to oversee it afterwards
Spanish organisations, Red Hat 2026 study
« el 34% realiza evaluaciones exhaustivas al principio del proyecto, pero reconoce dificultades para supervisar los riesgos posteriormente. »
Source: Étude Red Hat
-
32 %
share of organisations basing their strategy solely on initial assessments
Spanish organisations, Red Hat 2026 study
« Y un 32% basa su estrategia únicamente en evaluaciones iniciales. »
Source: Étude Red Hat
Classification
- Country
- Spain
- Company size
- All sizes
- Industries
- All industries, or industry not specified
- Companies
- Red Hat (study author)
- Keywords
- ai agents, ai governance, data traceability, digital sovereignty, software security, cio
Entry written by Marsanne Studio from the cited article, with the help of an AI model (claude-sonnet-5). The rights to the article and its illustrations belong to IT User. Entry translated from French by Claude Sonnet 5.5.