Published : 04/07/2026
Salesforce
Agentic AI in the public sector: anticipating needs and reshaping citizen experience
Alexander Brasaola
Salesforce Public Sector Director at Inetum
The relationship between citizens and public administration is entering a new phase, driven by the rise of agentic AI. This model, built on intelligent agents capable of understanding context, making decisions, and executing actions across complex processes, is shifting public services from reactive operations to continuous, personalized engagement. For public sector leaders, the question is no longer whether to adopt AI, but how to scale it in ways that are effective, responsible, and aligned with real-world needs. For years, digital transformation in government has focused on moving forms online, automating paperwork, and replicating in-person processes digitally. Yet for many citizens, the experience remains slow, fragmented, and unintuitive.
From reactive services to contextual engagement
While industries such as banking, retail, and mobility already deliver personalized, omnichannel experiences around the clock, many public institutions still operate in reactive modes. Citizens are expected to navigate systems, rather than systems adapting to citizens.
Agentic AI introduces a structural shift. Unlike traditional chatbots, it does not simply respond to predefined queries. It interprets context, draws on multiple data sources, and maintains conversational memory. In practice, this enables a more dynamic, adaptive interaction model that can evolve with each citizen’s situation. This capability matters in a public sector shaped by regulatory complexity, organizational silos, and multiple service channels. Whether applying for social benefits, managing licenses, or accessing healthcare services, citizens increasingly expect a seamless experience without repeating the same information across departments.
Citizen 360: building a unified view
A key enabler of this transformation is the concept of Citizen 360, a unified view of individuals that integrates data across systems, departments, and touchpoints. Historically, public administrations have relied on fragmented platforms and disconnected workflows, forcing citizens to repeatedly explain their situations. By contrast, modern platforms such as Salesforce Public Sector Solutions, Data Cloud, or Einstein allow organizations to maintain full interaction context and deliver consistent service regardless of channel, including phone, email, web portals, messaging apps, or in-person visits. This level of omnichannel continuity improves both citizen satisfaction and operational efficiency. It reduces duplication, accelerates case resolution, and simplifies collaboration across internal teams.
Moving from transactions to proactive services
One of the most significant shifts enabled by agentic AI is the transition from transaction- based models to proactive public services. Traditionally, citizens initiate processes: they identify a need, search for information, and apply for services. With agentic AI, administrations can anticipate needs based on life events such as the birth of a child, starting a business, relocating, or changes in health status. In these scenarios, intelligent agents can recommend relevant services, guide users through procedures, and automate parts of the process. This is especially impactful in social, healthcare, and employment contexts, where many individuals may not be aware of available benefits or how to access them. Beyond reducing administrative burden, this approach fundamentally simplifies how citizens engage with government, making services more accessible and equitable.
AI as a copilot for public servants
The organizational impact of AI is often viewed with concern, particularly regarding workforce implications. In practice, most public sector initiatives are not about replacement, but augmentation. Agentic AI is well-suited to handling repetitive, low-value tasks such as classifying requests, generating administrative summaries, or retrieving dispersed information. This frees public employees to focus on areas where human judgment remains critical, including complex decision-making, citizen support, and mediation. We are already seeing early examples. In regions such as Castilla-La Mancha, AI capabilities are being used to support citizen service agents through automatic transcription, summary generation, knowledge assistance, and workflow guidance for complex cases. The result is improved productivity without losing the human element.
The real challenges: data, governance, and trust
Despite rapid technological progress, the main barriers are not purely technical. Data quality remains a major constraint. Many public organizations still rely on legacy systems, incomplete datasets, and limited data governance models. Equally important is the need for transparency, traceability, and human oversight. AI systems cannot operate as opaque “black boxes” when making decisions that affect citizens. Governance frameworks must ensure explainability, accountability, and compliance with ethical standards. In practice, this means embedding control mechanisms, defining clear roles for human supervision, and maintaining citizen trust as a core priority.
From pilots to scalable impact
Many public administrations have already moved beyond experimentation. The current challenge lies in scaling successful pilots into operational models that deliver sustained value.
This requires more than deploying technology. It involves aligning governance, redesigning processes, and managing organizational change. In our experience, combining technological capability with a deep understanding of public sector dynamics is essential to achieving long-term impact. Agentic AI opens the door to a new way of designing public services: more personalized, more accessible, more proactive, and significantly less bureaucratic. The real transformation is not just about automating processes, but about rethinking how governments interact with citizens in a simpler, more contextual, and more efficient way.