AI ethics and governance in public infrastructure: Building trust, delivering value

It is often said that with great power comes great responsibility. As the adoption of Artificial Intelligence (AI) in the government sector grows and its impact on citizens’ lives expands, AI governance must be brought to the forefront to ensure the ethical, responsible, and effective use of the technology.  

AI is rapidly transforming government programs, including public infrastructure projects across the United States. As of January 2025, the Federal Agency Artificial Intelligence Use Case Inventory reported that 41 agencies had documented 2,133 AI use cases—up from 710 in 2023—with applications ranging from fraud detection to infrastructure maintenance and emergency response. DOTs accounted for 66 of these use cases. This transformation brings the promise of greater efficiency while also highlighting the need for ethical stewardship and accountability. The stakes are high: public trust, data privacy, and the effective use of taxpayer dollars all hang in the balance.

Government agencies are often responsible for handling sensitive data. To ensure effective oversight, managers of public infrastructure projects must use AI technologies responsibly. The imperative to protect data is set to intensify as the Trump administration’s $500 billion Stargate Project gains momentum—an ambitious initiative aimed at deepening collaboration with technology leaders and accelerating investment in national AI infrastructure.

At present, few laws and regulations govern the use of AI. However, new policies and rules are likely to emerge sooner rather than later. The rise of regulations will profoundly impact public infrastructure agencies. Stanford University’s Human-Centered Artificial Intelligence Report for 2025—widely regarded for its insights into the evolving landscape—found that the demand for AI skills in the U.S. is increasing, but roles focused on ethics, governance, and regulation account for just 0.02 percent of all job postings. This talent gap reveals a critical blind spot and underscores the urgent need for agencies to build internal governance capacity to responsibly harness the full potential of AI while maintaining public trust.

Ethical lapses—such as biased algorithms, opaque decision-making processes, or data misuse—can erode public confidence and trigger both public and regulatory backlash. Robust governance frameworks are essential to balance innovation with risk mitigation and to safeguard democratic values. Good governance not only ensures compliance but also helps avoid legal and reputational risks. The following six principles serve as a foundation for the responsible use of AI.

1. Transparency and explainability: Government agencies must ensure that AI-driven decisions are transparent, explainable, and open to scrutiny. This includes clear documentation of model design, decision logic, and data sources. The U.S. government’s April 2025 memorandum, “Accelerating Federal Use of AI through Innovation, Governance, and Public Trust,” marks a step toward the rapid and responsible implementation of AI at the federal level. Specifically, the memorandum requires agencies to develop public AI strategies, maintain accountability for spending, and foster transparency by sharing details of AI use cases and compliance plans. Agencies are also directed to appoint Chief AI Officers to oversee these efforts and ensure that AI adoption prioritizes accessibility, efficiency, and economic competitiveness while safeguarding civil rights and privacy.
2. Bias mitigation and fairness: To ensure fairness and reduce bias, AI systems must be designed using diverse and representative datasets. They should also incorporate third-party tools or techniques, such as reweighting or resampling, to address data imbalances. Regular, independent audits and transparent public reporting are essential to maintaining trust, especially in high-stakes areas such as public safety, hiring, and resource allocation, where biased outcomes could have significant negative consequences.
3. Risk assessment: Policy should mandate risk assessments to evaluate the impact of AI on public infrastructure, along with contingency plans for potential failures and mechanisms for stakeholder feedback to adapt to evolving challenges. Additionally, agencies must ensure that their AI tools are integrated with real-time monitoring capabilities to enhance safety, efficiency, and accountability throughout the project life cycle. Clear governance policies help prevent wasteful spending on poorly designed or misaligned initiatives, ensuring that taxpayer dollars are used effectively—and that agencies remain accountable for every public dollar assigned to them.
4. Data privacy and security: Protecting sensitive citizen data is non-negotiable. Agencies must comply with stringent data protection standards and conduct regular risk assessments to prevent misuse or breaches. To reinforce accountability, federal agencies are required—per government policy—to establish the roles of Chief AI Officers and an AI Governance Board. These structures ensure responsible oversight and clear stewardship of data while aligning security measures with evolving regulatory requirements and public expectations on an ongoing basis.
5. Accountability and oversight: Establishing clear governance structures—including ethics committees, independent oversight bodies, and public advisory boards—is essential for reviewing AI applications and addressing ethical dilemmas in public infrastructure. Integrating mechanisms for stakeholder and public feedback into these governance processes helps ensure that AI systems align with the community’s values and needs, fostering transparency. Ultimately, effective governance mitigates risks, unlocks innovation, and lays the foundation for long-term success.
6. Public-Private Partnerships (PPP): By fostering collaboration with private-sector innovators, academia, and civil society, public infrastructure agencies gain access to technical expertise and diverse perspectives—an approach also recommended by organizations such as World Economic Forum. The PPP model not only helps bridge resource gaps but also ensures that infrastructure projects leverage the latest technological advances while maintaining a strong commitment to the public interest and ethical standards. Furthermore, actively engaging the public in both the design and oversight of AI systems ensures that solutions remain responsive to the community’s diverse needs.

The promise of AI in public infrastructure is clear: smarter cities, safer roads, more responsive services, and better use of taxpayer dollars. But realizing this promise requires more than technical prowess—it demands a culture of ethical leadership, robust governance, and an unwavering focus on the public good. As the U.S. embarks on unprecedented investments in AI infrastructure, wisdom rooted in ethics and driven by public value will be the cornerstone of sustainable progress.  

Equally important is the impact of AI on the public workforce. Fear of job displacement can slow innovation, but AI is not a substitute for human insight—it is a tool that can ease the growing demands on government teams. When deployed thoughtfully, it enables staff to concentrate on what matters most: delivering exceptional service, fostering safer communities, and maintaining resilient infrastructure. 

Public infrastructure leaders now stand at a pivotal moment—not only to meet emerging governance standards but to define them. By leading with ethics, transparency, and collaboration, they can set a national benchmark for responsible AI—one that delivers lasting value for all citizens.