Artificial intelligence (AI) is increasingly shaping conversations about modernization across government technology. Its promise of automation, predictive analytics, and operational efficiency is widely discussed, even as questions persist around trust, security, explainability, and control.¹

For government agencies, the challenge is not whether AI can deliver value, but whether it can be applied to environments where transparency, statutory compliance, and public accountability are non negotiable. Systems that support elections, oversight, and regulatory compliance must withstand scrutiny, produce defensible outcomes, and preserve human judgment – requirements that complicate AI-first adoption.

This white paper argues that Data Intelligence, rather than AI-first adoption, offers a more practical and trustworthy path forward for public sector systems. Data Intelligence emphasizes disciplined data governance, explainable analytics, and human oversight, with AI applied selectively as a supporting capability rather than a replacement for professional judgment.

Drawing on enterprise research, U.S. public sector studies, and emerging government use cases, this paper examines the opportunity and risks associated with AI, explains why Data Intelligence is better aligned with state government software systems, and outlines a responsible
approach to innovation that strengthens public trust.

THE OPPORTUNITY AND LIMITS OF AI IN GOVERNMENT

Interest in AI continues to accelerate across both the private and public sectors. Enterprise research highlights its potential value while also revealing significant readiness gaps.

An IBM study found that while approximately 63 percent of enterprise data is considered valuable for AI, only about 7 percent is actively used in AI systems today, underscoring persistent challenges around data preparation, governance, and trust.²

Within U.S. government systems, federal guidance has consistently emphasized that AI’s benefits depend on strong data management, risk controls, and institutional oversight. The National Institute of Standards and Technology (NIST) notes that AI systems used in high-impact environments must be governed to ensure reliability, explainability, fairness, and accountability throughout their lifecycle.³

These findings reflect a central reality for government systems. AI’s potential is real, but its effectiveness depends on the quality, governance, and integrity of the underlying data, and on the ability to explain and defend outcomes.

THE CENTRAL TENSION: INNOVATION VS. TRUST

Government leaders increasingly recognize that delaying modernization carries its own risks. At the same time, they cannot adopt AI without adequate safeguards.

Research indicates that nearly 70 percent of executives believe AI’s productivity gains are significant enough to justify some level of risk, even amid uncertainty about governance and implementation.⁴

Yet trust remains fragile. As one enterprise analysis observes, only a small minority of organizations fully trust AI agents to autonomously run core business processes.⁵

For government agencies, where decisions are subject to legal challenge, public scrutiny, and democratic accountability, this tension is amplified. Innovation must deliver tangible benefits, such as faster compliance reviews, improved oversight, and earlier detection of risk, without sacrificing explainability, auditability, or human control.

FROM AI TO DATA INTELLIGENCE: A MORE PRACTICAL STRATEGY

Rather than treating AI as a standalone solution, many government organizations are increasingly gravitating toward Data Intelligence as a governing strategy.

Data Intelligence refers to the structured use of data, supported by analytics, rules-based logic, and, where appropriate, machine learning, to generate insights that are explainable, auditable, and aligned with human decision making.

In this model:

• AI is a capability, not the strategy itself.
• Data Intelligence is the framework that determines when, where, and how advanced techniques should be applied.

Without a Data Intelligence strategy, AI can increase risk by amplifying data quality issues, obscuring decision logic, or creating governance gaps. Within a Data Intelligence framework, AI can enhance value by accelerating pattern detection, prioritizing human review, and surfacing insights that would otherwise remain hidden.

GOVERNMENTS ARE EXPERIMENTING, CAREFULLY

AI adoption in government is no longer theoretical, but it remains measured.

According to the Brookings Institution, documented federal AI use cases grew from 710 in 2023 to more than 1,750 in 2024, spanning agencies such as the U.S. Patent and Trademark Office and the Transportation Security Administration.⁶

At the state and local level, initiatives such as e.Republic’s AI 50 highlight practical examples of agencies using technology to improve efficiency, transparency, and service delivery.⁷

Notably, many of these efforts begin with lower risk applications, such as anomaly detection, pattern analysis, and workflow prioritization, before moving toward more deeply embedded intelligence. This phased approach reflects an implicit shift toward Data Intelligence-first adoption.

GOVERNANCE AND RISK ARE NOT SECONDARY CONCERNS

Research consistently shows that AI systems lacking transparency and oversight can erode trust, even when they improve efficiency.

Academic studies suggest that while AI-enabled systems may streamline processes, they can reduce citizens’ sense of control when decisions are not explainable or contestable, reinforcing the need for human oversight and clear governance frameworks.⁸

U.S. oversight bodies have raised similar concerns. The Government Accountability Office (GAO) has warned that agencies deploying AI without strong data governance and risk management practices may expose themselves to operational, legal, and reputational risks, particularly when systems rely on incomplete or poor-quality data.⁹

These risks are especially pronounced in domains such as elections, ethics reporting, and business registration, where public confidence is inseparable from system integrity.

HOW MODERN GOVERNMENT PLATFORMS ENABLE RESPONSIBLE DATA INTELLIGENCE

Responsible Data Intelligence is not enabled by algorithms alone. It depends on the software environment in which data is governed, analyzed, and acted upon.

Modern government platforms increasingly play a critical role in supporting Data Intelligence by providing structured, managed environments where governance, security, and oversight are embedded into the system itself.

These platforms support Data Intelligence in several important ways:

Governance at the platform level: Well-architected systems centralize security controls, access management, and data standards, allowing intelligence capabilities to operate within clearly defined guardrails rather than as isolated or experimental tools.

Consistency and auditability: When intelligence functions are embedded within a common platform, insights are generated using consistent logic, data definitions, and controls. This supports audit trails, oversight review, and defensible decision-making.

Evolution without disruption: Modern platforms allow analytical techniques, validation rules, and governance controls to evolve over time without requiring agencies to rebuild systems or disrupt operations. This enables innovation while preserving stability.

Separation of insight and authority: Platform-based approaches reinforce an important distinction: intelligence surfaces signals and priorities, while authority remains with human decision makers operating within statutory frameworks.

Enterprise research suggests that organizations are increasingly favoring platform-centered approaches to advanced analytics, prioritizing reliability, explainability, and governance over bespoke
or experimental implementations.⁵

A RESPONSIBLE DATA INTELLIGENCE FRAMEWORK FOR GOVERNMENT

Based on industry research and public sector experience, a responsible Data Intelligence approach
should include:

Human-in-the-loop decision-making: Insights should support staff judgment, not replace statutory authority or professional discretion.

Explainability and auditability: Outputs must be understandable to auditors, oversight bodies, and
the public.

Governance by design: Security, ethical use policies, and data quality standards should be established before intelligence capabilities are scaled.

Incremental adoption: Begin with defined insights and anomaly detection, then progress toward
embedded intelligence as trust and governance mature.

CONCLUSION: INNOVATION THAT REINFORCES PUBLIC TRUST

AI will continue to shape discussions about government modernization, but lasting progress depends on how it is governed. Data Intelligence provides a practical framework for innovation that aligns with public sector realities, balancing efficiency gains with transparency, accountability, and trust.

When implemented responsibly, Data Intelligence strengthens the integrity of public systems, improves compliance outcomes, and enables government officials to focus on higher value oversight and service delivery. When governance lags behind innovation, the risks extend beyond technology and into public confidence itself.

---

---