Indra Group has become the first Spanish company to receive AENOR’s “Responsible AI Technology based on Microsoft tools” certification, validating the way it designs, deploys and operates selected AI agents built with Microsoft technology.
The certification was presented in Madrid on July 13, with Indra Group, AENOR and Microsoft Spain represented at the event. As reported by Indra Group, the assessment is intended to test real operational controls rather than simply endorse a set of AI principles or a policy document.
AENOR’s scheme focuses on AI agents developed with Microsoft tools and assessed against Indra’s internal agent-development methodology. The controls cover traceability, safeguards, data quality, cost management, technical configuration, version control, testing and evidence collection across an agent’s lifecycle.
That distinction matters. Enterprise AI governance is increasingly less about the underlying model alone and more about the system wrapped around it: what data it can access, who can change its instructions, how its output is monitored, and whether an organization can show evidence that those controls were applied.
Microsoft and AENOR began collaborating on an ethical and responsible AI certification framework in September 2024. Microsoft said at the time that the work would draw on international standards, including ISO/IEC 42001, as well as Microsoft’s responsible-AI principles and European AI regulation.
That is a useful limitation to keep in mind for IT teams reading the announcement. A certification of particular agents and their operating processes is not a blanket assurance that all generative-AI output is accurate, safe or suitable for unsupervised decisions. It is evidence that the certified use cases have documented controls and an auditable management process.
The practical message for Windows and Microsoft 365 administrators is straightforward: treat agents as managed production systems. Define approved data sources, retain configuration and test records, establish owners, log changes, set spending controls, and keep humans in the loop for compliance or HR workflows.
Indra’s certification may offer a model for organizations looking to formalize those controls, but it currently applies only to the two named Copilot Studio agents.
The certification was presented in Madrid on July 13, with Indra Group, AENOR and Microsoft Spain represented at the event. As reported by Indra Group, the assessment is intended to test real operational controls rather than simply endorse a set of AI principles or a policy document.
What AENOR assessed
AENOR’s scheme focuses on AI agents developed with Microsoft tools and assessed against Indra’s internal agent-development methodology. The controls cover traceability, safeguards, data quality, cost management, technical configuration, version control, testing and evidence collection across an agent’s lifecycle.That distinction matters. Enterprise AI governance is increasingly less about the underlying model alone and more about the system wrapped around it: what data it can access, who can change its instructions, how its output is monitored, and whether an organization can show evidence that those controls were applied.
Microsoft and AENOR began collaborating on an ethical and responsible AI certification framework in September 2024. Microsoft said at the time that the work would draw on international standards, including ISO/IEC 42001, as well as Microsoft’s responsible-AI principles and European AI regulation.
Two Copilot Studio agents in scope
The certification does not appear to cover every AI deployment across Indra Group. It applies to two specific agent use cases built with Microsoft Copilot Studio:- An onboarding assistant for employees joining the group.
- A regulatory-compliance supervisor supporting verification and continuous improvement for Indra’s ISO/IEC 20000 management system.
That is a useful limitation to keep in mind for IT teams reading the announcement. A certification of particular agents and their operating processes is not a blanket assurance that all generative-AI output is accurate, safe or suitable for unsupervised decisions. It is evidence that the certified use cases have documented controls and an auditable management process.
Why it matters for Microsoft shops
For organizations rolling out Copilot Studio agents, the announcement is another sign that governance is becoming a deliverable alongside the bot itself. Microsoft’s responsible-AI framework emphasizes fairness, reliability and safety, privacy and security, inclusiveness, transparency and accountability, but customers remain responsible for how they configure and operate their own applications.The practical message for Windows and Microsoft 365 administrators is straightforward: treat agents as managed production systems. Define approved data sources, retain configuration and test records, establish owners, log changes, set spending controls, and keep humans in the loop for compliance or HR workflows.
Indra’s certification may offer a model for organizations looking to formalize those controls, but it currently applies only to the two named Copilot Studio agents.
References
- Primary source: Atalayar
Published: 2026-07-13T15:00:00+00:00
Indra Group, the first Spanish company to be certified by AENOR for the responsible use of artificial intelligence
Artificial intelligence can improve the onboarding processes for professionals joining large companies, as well as the monitoring and improvement of manageme...www.atalayar.com - Related coverage: indragroup.com
Icon Isotype
Indra Group ha obtenido la primera certificación de AENOR de “Tecnología de IA Responsable basada en herramientas Microsoft”, que certifica la aplicación efectiva de los principios de IA responsable en agentes de IA evaluados; siendo, por ello un reconocimiento al uso responsable de las...www.indragroup.com - Official source: learn.microsoft.com
Información general sobre inteligencia artificial - Microsoft Service Assurance | Microsoft Learn
Más información sobre la inteligencia artificial en Microsoft Online Serviceslearn.microsoft.com
