How EY built an enterprise-scale agentic AI operating system

Furthering the global EY journey to AI transformation, EY.ai EYQ was deployed to more than 300,000 professionals, powering secure enterprise chat, domain assistants, governed prompt tooling and safe generative AI (GenAI) experimentation across all service lines. The firm rolled out Microsoft 365 Copilot at scale and integrated AI into real business workflows across Tax, Assurance, Consulting and internal functions. The EY organization was no longer experimenting with GenAI — it was operating it at global enterprise scale.

But as the ambition of the organization grew, so did the complexity of the challenge. Its goal of reaching US$100bn in revenue was closely tied to AI. To achieve this, the organization needed to protect its existing business, disrupt its own models and grow entirely new ones. Leadership recognized that CEOs worldwide were struggling to scale AI pilots and needed a trusted guide. EY teams understood early that no organization, including itself, could scale AI alone. AI was no longer just a productivity tool; it was a growth imperative.

This raised a deeper strategic question. As EY teams began developing agents, workflows, copilots and custom models, fragmentation emerged as a new challenge. AI use cases now required multistep autonomy rather than single-step assistance. Real value depended on AI-ready data, lineage, consent governance and interoperability across EY Fabric, Microsoft technologies, client systems and partner ecosystems.

Meanwhile, 400,000 EY people needed to shift from “using AI” to working alongside fleets of agents. Regulatory expectations from bodies such as the EU and G7 heightened the need for Responsible AI built into every system.

The answer began with the realization that employees didn’t need another tool, model or assistant — but rather a comprehensive operating system for agentic work across the enterprise. EY.ai EYQ provided the foundation, serving as the original enterprise GenAI ecosystem and the most robust governed GenAI environment the organization had developed. The next step was to evolve upward: unifying intelligence, orchestration, data, workflows, governance and domain experience into one coherent system.

This new operating system needed to support autonomous, multistep workflows, operate under strict Responsible AI and regulatory expectations, scale to 400,000+ people, integrate with EY business operations and extend through the EY Partner Ecosystem.

“We had to adopt a strategy of not going alone,” explains Mark Luquire, EY Global Microsoft Alliance Co-innovation Leader. “We needed to use our alliance relationships across all of our technology partners.”

The team has identified three foundational capabilities essential for success:

Unified intelligence layer

EY teams needed a single way to access, tune, govern and deploy foundation models and agentic reasoning capabilities. This required a centralized model catalog, tuning and distillation pipelines, guardrails, multimodal intelligence and planning and tool use frameworks.

Unified orchestration and workflow layer

EY needed an orchestration layer capable of running autonomous, end-to-end workflows across Microsoft 365, EY Fabric, and the broader EY technology landscape. Agents needed to surface directly inside everyday tools — email, documents, meetings — while supporting lifecycle management, security, identity and multi-agent coordination at scale.

Unified data and trust foundation (AI-ready data)

Agentic systems are only as strong as their data. EY professionals required governed, permissioned, lineage rich, and compliant data, tied together across EY Fabric, client systems and cloud ecosystems. The EY.ai Data Marketplace provides this trusted foundation as the engine that connected AI-ready data to directly improve model performance, agent effectiveness, and ultimately business value.

What EY built: the EY.ai Agentic Platform

EY teams designed a platform that is extensible, portable, multi-technology and built to operate across both Microsoft and NVIDIA environments — for internal and client applications.

Layer 1 — Intelligence

NVIDIA powers graphics processing units (GPUs), simulation, NVIDIA inference microservices (NIMs), NeMo Guardrails and Foundry for training and distillation. EY layers domain-specific intelligence for each service line and uses Omniverse and simulation environments for physical AI and robotics.

Layer 2 — Orchestration

Microsoft powers the orchestration plane through Azure Microsoft Foundry’s 11,000 model catalog, Copilot as the front door to AI, Microsoft Fabric for data unification and Copilot Studio for enterprise agent creation. EY.ai EYQ evolves into the enterprise marketplace where agents are created, discovered and consumed.

Layer 3 — Data, trust and governance

Microsoft Azure and other technologies power the AI data plane. The EY.ai Data Marketplace and EY assurance teams provide governance, Responsible AI frameworks, auditability and regulatory alignment across the entire stack.

The ecosystem that accelerates it

Microsoft is one of EY’s largest cloud partners, and EY also works closely with NVIDIA. NVIDIA Omniverse is available on Azure, Microsoft’s cloud platform, and EY’s role is to orchestrate this ecosystem to deliver seamless solutions for clients.

This agentic AI platform exemplifies how advanced technologies can transform professional services by enhancing efficiency and empowering teams. By integrating AI into critical workflows, EY is addressing persistent challenges in the industry, such as the need for rapid data analysis and compliance management, while promoting high standards of quality and reliability. In just nine months, EY.ai has accelerated agentic innovation across EY, driving both the experimentation and development of over 50,000 agents.

This platform strategy is a pivotal solution to resource limitations, expanding the capacity of our workforce beyond traditional constraints. It enables EY to scale human expertise faster than conventional hiring and training could support, directly contributing to the organizational ambition of market-leading growth. This was a critical next step in becoming a frontier with an AI-enabled workforce. EY teams are now better equipped to orchestrate humans and AI agents in tandem, enhancing our strategic workforce and scaling human capacity to meet growing demands. This evolution positions EY to continue advancing toward an AI-enabled operating model where autonomy and human ingenuity work in concert.

This initiative also responds to the evolving demands of the workforce, where the integration of AI serves to augment human capabilities rather than replace them. With over 80% of EY professionals using EY.ai EYQ and more than 80% completing foundational AI training, this is helping our workforce remain focused on strategic insights and complex decision-making. This approach supports continuity in service delivery and helps organizations meet client expectations while prioritizing employee well-being. This amplification of capability and capacity is critical to helping EY achieve its ambition of market-leading growth.

Importantly, EY has adopted a thoughtful and responsible approach to AI innovation. By emphasizing ethical governance, security and rigorous training, the organization is building confidence and technical knowledge within its workforce. AI systems are initially tested in controlled environments, allowing for the gradual evolution of safety protocols and operational readiness alongside technological advancements.

The initiative reflects a broader trend in how organizations are embracing AI. Rather than merely seeking immediate efficiency gains, the EY strategy focuses on long-term capability development. Leveraging AI in EY’s training and hiring practices has enabled it to scale human expertise faster than would have been possible with traditional approaches. With an AI-enabled workforce, EY is now better equipped to see the full benefit of humans and AI agents working in concert to enhance its strategic workforce. The agentic AI platform is viewed as a strategic asset that necessitates ongoing investment, collaboration, and continuous learning. With 2 million learning hours consumed, and 300 degrees awarded in partnership with Hult International Business School, EY is committed to fostering a culture of continuous improvement.

Looking ahead, EY plans to refine its AI capabilities further, expand into new service areas, and prepare for broader deployment across its global operations. While it is still early to measure the full impact, this initiative aims to position EY as a leader in the responsible adoption of AI, fostering a new era of professional services that balances autonomy with human ingenuity, innovation with ethical considerations, productivity with resilience, and experimentation with operational excellence.

Through its collaboration with technology partners like Microsoft and NVIDIA, EY is establishing a practical framework for how professional services can leverage AI responsibly, so that innovation drives sustainable transformation while maintaining a strong commitment to governance and client trust.