Balancing innovation, risk and public trust

“Government agencies are not racing competitors. They're balancing modernization against responsibility, security and public trust. AI changes how agencies modernize, but success still depends on leadership, governance and a relentless focus on constituent outcomes."

State and local governments are navigating a complex mix of aging technology, evolving constituent expectations, workforce constraints and increasing demands for operational efficiency. As agencies balance innovation with responsibility, AI is creating new opportunities to modernize faster while maintaining public trust. We spoke with Virginia Williams, a leader with extensive experience in state and local government consulting, about how agencies can modernize successfully while managing risk and delivering measurable public value.

What is driving modernization in state and local government today?

Modernization isn't new for government agencies. Many organizations have been dealing with aging systems, legacy applications and even mainframe environments for years.

What's changing is the urgency. Agencies are facing increasing expectations from constituents who want the same level of digital service they experience in their personal lives. At the same time, they're dealing with staffing shortages, budget constraints and growing pressure to improve operational efficiency.

The challenge isn't deciding whether modernization is necessary. Most agencies already know it is. The challenge is determining how to modernize in a way that delivers meaningful outcomes without increasing operational or political risk while also demonstrating responsible stewardship of taxpayer dollars. Government leaders are accountable not only for delivering better services, but for ensuring modernization investments create measurable public value.

For government leaders, modernization isn't really a technology conversation. It's a public value conversation.

How is AI changing the way agencies approach modernization?

I often describe it as two conversations happening at the same time. The first is using AI to modernize systems. The second is building AI into the future-state environment.

Today we're seeing agencies use AI to accelerate activities such as requirements gathering, discovery, prototyping and use case generation. Those are all parts of the modernization process that historically required significant manual effort.

What's becoming increasingly interesting is that AI is starting to influence decisions about the target technology environment itself. Historically, low-code and SaaS platforms were especially attractive because government agencies often struggled to recruit and retain highly specialized technical talent. Today, AI is reducing some of those barriers by making software development more accessible and efficient.

As a result, agencies are beginning to reassess long-held assumptions about platform strategy, vendor lock-in and the cost of custom development. At the same time, the security, accessibility and scalability provided by established platforms remain highly attractive. The organizations moving fastest are evaluating how to balance both worlds: the flexibility enabled by AI-driven development and the governance advantages of mature enterprise platforms.

Cost predictability is becoming part of that conversation as well. As software providers increasingly shift toward consumption-based pricing models, agencies must evaluate not only technical capabilities but also how those costs align with budget cycles that are often established years in advance.

Government agencies are investing in AI, but adoption looks different from that in the private sector. Why?

Government operates under a very different set of incentives. Commercial organizations are often motivated by competitive pressure. They need to reduce costs, innovate faster or gain market advantage. Government agencies are focused on service quality, taxpayer value, compliance and public accountability. The race to get there doesn't really exist in the same way.

That reality shapes how agencies adopt AI. Many organizations are beginning their AI journey by leveraging capabilities already available within technologies they use today. That may include AI functionality embedded in platforms such as Salesforce, ServiceNow or Microsoft, but it also extends to specialized government applications where AI can automate activities such as invoice processing, document management or constituent interactions.

By building on existing investments, agencies can introduce AI within established workflows while reducing implementation risk and procurement complexity.

Procurement cycles also play a significant role. Agencies may be working with budgets that were approved years before today's AI capabilities even existed. It's often easier to enable functionality within an existing platform than to launch an entirely new procurement effort.

What challenges prevent agencies from modernizing successfully?

One of the biggest challenges is what I call the chase for perfection. Government agencies don't always receive funding for major technology initiatives on a regular basis. When funding becomes available, there's often pressure to solve every problem at once and design the perfect future state before moving forward. The reality is that perfection is rarely achievable.

Many agencies are trying to envision exactly what they want years before they ever experience the capabilities of a modern platform. That can lead to complexity, delays and stalled projects.

The organizations making the most progress are focusing on manageable improvements that deliver value quickly. They're willing to learn, adjust and continue improving rather than spend endless cycles designing and developing the perfect end state before moving forward.

How important are workforce challenges in the modernization conversation?

They're extremely important. Public-sector organizations continue to face challenges recruiting and retaining technical talent, particularly in areas like cloud engineering and AI. At the same time, many agencies are being asked to deliver more services without adding staff. That's why many leaders view AI as a force multiplier.

The goal isn't replacing people. It's helping organizations operate more effectively with the resources they have. AI can automate repetitive work, reduce administrative burden and allow employees to focus on higher-value activities that directly support constituents.

For many agencies, workforce constraints are a more immediate driver than the technology itself.

What role does governance play in AI-enabled modernization?

Governance is becoming even more important. Historically, modernization programs were often measured by project milestones, completed requirements or successful deployments. Those metrics are still important, but they don't necessarily tell you whether modernization is improving outcomes.

The agencies making the most progress are measuring modernization through outcomes rather than delivery milestones. They focus on indicators such as reduced case-processing times, improved constituent experiences, operational efficiency and taxpayer value.

Contract structures also deserve careful consideration. AI should enable modernization partners to deliver value faster and more efficiently. Traditional time-and-materials or staff-augmentation models can unintentionally create the opposite incentive because they reward effort rather than outcomes. As AI continues to improve productivity, agencies and their partners will increasingly benefit from models that align compensation with measurable performance improvements rather than hours worked.

That's where modernization shifts from being a project to being a performance discipline.

What are some examples of modernization success you're seeing today?

Many state agencies have made significant progress modernizing infrastructure, moving applications to the cloud and reducing dependence on legacy systems. In some cases, agencies have successfully retired mainframe environments and adopted modern architectures that create greater flexibility for future innovation.

For example, in California, agencies have made significant progress in modernizing infrastructure, moving applications to the cloud and reducing dependence on legacy systems. This includes the continued modernization of mission-critical platforms such as the Case Management, Information and Payrolling System (CMIPS), which supports California’s In-Home Supportive Services program. Ongoing enhancements have significantly expanded the system’s capacity and performance, helping California reliably process a growing volume of caregiver timesheets and payments while creating a more scalable foundation for the future.

What advice would you give government leaders beginning their modernization journey?

Focus on meaningful outcomes. Identify a problem that matters. Choose a use case that can deliver visible value. Build momentum through early wins.

Modernization is an organizational change effort. People need time to adapt to new processes and new ways of working. Early successes help build confidence and create support for larger changes over time.

I'd also emphasize the importance of leadership. The agencies making the greatest progress have a visionary leader who is willing to champion change, encourage innovation and create momentum across the organization.

Technology matters, but leadership is what ultimately determines whether modernization becomes a sustained capability or another initiative that loses momentum.

The agencies that make the greatest long-term progress won't necessarily be the first to adopt every new AI capability. They'll be the ones that use AI responsibly to improve constituent outcomes, operational performance and taxpayer value continuously.