Reimagining the Center of Government for the age of intelligence

By Nizar Hneini and Hugo Carreira

Around the world, governments are moving swiftly to harness the potential of artificial intelligence (AI). New digital authorities and data offices are emerging at record speed, and national strategies continue to multiply. Yet despite this flurry of activity, the machinery of government often operates much as it did decades ago. Structures that once ensured stability now limit agility. Hierarchies and fixed decision flows remain, designed for a slower era. The challenge is not technology itself, but that systems have not evolved to prioritize intelligence. We examine where the solution lies – and propose a new way for governments to connect strategy, data and delivery into a single intelligent system.

According to the OECD Digital Government Index (2023), only 15 percent of countries have cross-ministerial data integration mechanisms to steer national strategies. Similarly, the UN E-Government Survey (2024) shows that while 80 percent of member states now have digital or AI strategies, fewer than a quarter have aligned them across portfolios.

If digitalizing services defined the last decade, the next must transform how government itself operates. The Center of Government must evolve from coordinating information to orchestrating intelligence – linking insight, foresight and delivery.

An AI-enabled operating model

AI-enabled operating models go far beyond technology: They redefine how government thinks, decides and delivers.

Roland Berger’s AI-enabled Government Operating Model (AIGOM) is a practical framework that integrates five dimensions – strategy, governance, organization, technology and performance – to enable governments to govern through intelligence. Its components are:

• Strategic intelligence: Using data to guide decisions
• Adaptive governance: Creating flexible management structures
• A capability network: Sharing expert resources across ministries
• Anticipatory service delivery: Applying predictive, AI-based services
• Outcome intelligence: Measuring the real impact of actions

How the model works

In an AI-enabled operating model introduce a number of key changes in how governments operate. Strategy becomes dynamic as predictive analytics transform static plans into adaptive portfolios where priorities and resources shift in real time. Governance becomes adaptive, with intelligence councils bringing together policy, data and delivery. The organizational setup becomes a capability network, making shared expertise accessible across ministries. Services become anticipatory: AI predicts citizen needs and enables proactive responses, while unified data standards and interoperability create a single governmental “nervous system.” Finally, performance becomes intelligent, with real-time dashboards tracking budgets, programs and outcomes, turning data into continuous feedback for decision-makers. Collectively, these shifts redefine the Center of Government as a dynamic, learning institution that orchestrates national intelligence rather than merely enforcing compliance.

Case study: A national data platform for the United Arab Emirates

Under We the UAE 2031, the Prime Minister’s Office launched a national data platform linking sustainability, digital economy and social development missions across ministries. Dashboards now connect policy outcomes to resource allocation, enabling near real-time adjustments. According to TDRA (2024), data integration across these portfolios has reduced duplication by more than 20 percent – a clear demonstration of AIGOM principles in action, where intelligence becomes the mechanism of coordination.

Case study: Singapore’s Smart Nation initiative

Singapore’s Smart Nation strategy offers one of the world’s most mature examples of AI-enabled governance. Through platforms such as CODEX (Core Operations, Development Environment and eXchange) and Moments of Life, the government integrates data across ministries to deliver life-journey-based services, from birth registration to elder care. AI-driven analytics continuously assess service performance, enabling proactive adjustments and personalized interventions. Singapore demonstrates how anticipatory service delivery – a cornerstone of the AI-enabled operating model – turns government into a real-time orchestrator of citizen experience and national outcomes.

Case study: Estonia’s X-Road and AI governance framework

Estonia’s X-Road platform connects more than 1,000 public and private institutions through a secure, interoperable data exchange layer – the digital backbone of the Estonian state. Building on this foundation, the government’s AI Strategy 2030 embeds machine learning in public service automation, fraud detection and predictive policymaking. Today, over 80 percent of government services are fully digital and data-driven, enabling near real-time decisions across ministries. This exemplifies how institutional intelligence, not just digital efficiency, can transform government into a continuously learning system.

Governing through intelligence

In the industrial era, governments coordinated through plans, committees and memoranda. In the age of intelligence, learning and adaptation must guide governance. AI now allows leaders to understand system-wide behavior and act on it instantly. When investment in clean transportation reduces emissions but increases electricity demand, AI models reveal the trade-off immediately. When employment programs underperform in a specific region, predictive tools flag the risk and recommend preventive action.

Historically, Centers of Government existed to enforce consistency and control. In the age of intelligence, their role is to facilitate large-scale learning. This shift requires flatter structures, open access to information and collaboration across teams. It replaces secrecy with transparency and process with outcomes, focusing on results rather than rigid procedures. Governments that adapt will shape the future; those that do not will be shaped by it.

An intelligent Center of Government does not replace leadership judgment – it strengthens it, providing foresight, context and confidence amid uncertainty. Governance evolves from static coordination to real-time orchestration.

At Roland Berger, we believe that institutional intelligence – the ability of governments to connect missions, data and decisions into a single adaptive system – will define national competitiveness in the era of AI. The AI-Enabled Government Operating Model offers a blueprint for this transformation, showing how leaders can design governance that learns as it delivers, integrating technology, foresight and performance into a unified, continuously learning architecture.

Signals and implications for leaders

Global benchmarks reinforce the shift toward intelligence-driven governance. The OECD Future of Government (2025) report finds that countries with integrated AI and data-governance frameworks deliver policy outcomes up to 30 percent faster than those operating in silos. The World Economic Forum (2024) highlights cross-sector coordination and real-time data sharing as critical to transformation success, while the UN E-Government Survey (2024) notes the convergence of mission-based governance models among leading digital nations. Together, these findings confirm that intelligence-driven, multi-sector governance is becoming the global standard for achieving national impact.

The way forward

AI now forms the foundation of governance, connecting strategy, delivery and measurable outcomes. According to the World Economic Forum (2025), governments lose up to 40 percent of national project value to coordination inefficiencies. Our AI-enabled operating model helps recover this value through intelligence-driven governance.

Government that embrace this shift will not only operate more efficiently but will also learn, adapt and deliver lasting impact. The opportunity is immediate and the imperative clear: re-imagine the Center of Government as the hub of national intelligence.

The task ahead is not simply to make government digital, but to make it think - coherently, ethically and at the speed of its citizens’ expectations.