Quick Buzz Feed
Responsible AI starts with who is in the data, who is at the table, whose needs shape the outcome, and who is responsible when it falls short.
Rural communities, while possessing strong internal resources like social connectedness and local knowledge, frequently encounter obstacles such as lower population density, wider geographic dispersion, and limited access to essential services and infrastructure. In these environments, AI decisions can quickly impact multiple domains, emphasizing the need for locally relevant context and community oversight. The increasing adoption of AI impacts not only specific tools but also how rural needs are represented in data, who participates in AI governance, and how the benefits and burdens of AI are distributed, raising concerns about whether AI systems adequately address rural conditions and experiences.
Rural invisibility in AI systems occurs when rural communities are underrepresented in the data, assumptions, design, validation, and governance of AI. This makes it difficult to recognize rural needs and detect potential harms. Research indicates that rural AI studies are limited, models often underperform in rural settings, and the consequences of these failures are rarely examined where they are most felt. This invisibility means AI systems might be built on assumptions that do not reflect rural realities, leading to overlooked needs in resource and service allocation, and potentially causing missed diagnoses, misallocation of resources, and increased strain on rural healthcare providers due to fragmented records, thin staffing, and delayed care pathways.
Current AI governance frameworks, while providing a strong foundation, often fall short in offering practical guidance for implementing responsible AI principles in rural environments. They do not adequately account for disparities in data availability, institutional capacity, and technical expertise common in rural areas, nor do they typically mandate testing across small or geographically distinct populations, allowing rural-specific issues to go unnoticed. Furthermore, these frameworks often lack specific mechanisms for incorporating local knowledge and community perspectives, leading to algorithmic and broader rural invisibility in AI decision-making. Federal policy, led by agencies like OSTP, OMB, HHS, USDA, and ED, could provide crucial support by signaling priorities, shaping governance expectations, strengthening rural data infrastructure, and developing practical examples to address these gaps and foster more equitable AI implementation.
Rural proofing is a systematic process to ensure policies, tools, and investments genuinely reflect rural realities, prevent unintended harms, and promote fair outcomes for rural communities. Applied to AI, it involves making rural conditions visible throughout system design, data collection, deployment, and oversight. This includes defining clear use cases, engaging communities in AI decisions, transparently explaining system functions and limitations, and regularly reviewing performance to ensure equitable results, especially for small or low-volume populations. Rural proofing mandates that rural context and strengths are central to AI design and evaluation, not an afterthought. It also emphasizes practical, accountable governance with plain-language documentation and clear processes for identifying and addressing AI failures in rural settings, recognizing the unique constraints of rural systems like limited staff and budgets.
Addressing the invisibility of rural communities in AI systems requires coordinated national attention and action, including integrating rural proofing into AI governance. While national frameworks guide states, states can proactively build their own pathways by aligning with existing frameworks, piloting new approaches in key areas, and strengthening internal capacities. This memo outlines a specific plan for Kansas, recognizing its strong interest in ensuring AI systems effectively serve its rural communities across sectors like healthcare, education, and public services.
Kansas, as a predominantly rural state, is uniquely positioned to become a national leader in rural AI governance. The state can leverage its existing rural health infrastructure and engage local stakeholders to test practical rural-proofing approaches. This includes integrating rural context into the Kansas Legislative Artificial Intelligence Task Force, aligning technology investments with rural AI proofing principles via programs like CMSâs Rural Health Transformation Program, and ensuring agencies like KDHE, KDADS, and DCF translate broad AI governance principles into practical oversight for rural health and social systems. The implementation should be phased, starting with foundational actions and gradually building a coordinated, statewide approach.
The Kansas Legislative Artificial Intelligence Task Force, and any subsequent state-level task forces, should explicitly incorporate rural context, AI use, and governance into their core mandate, membership, and workplan. This involves defining the scope to include reviewing AI use in rural areas, integrating rural and frontline perspectives in AI decisions, and issuing guidance on procurement and accountability for rural health and social systems. Coordination should be led by the Office of Information Technology Services (OITS) with statewide governance direction from the Information Technology Executive Council (ITEC), while service agencies like KDHE, KDADS, and DCF identify where AI impacts health and social service access, establishing a clear governance structure tailored to rural conditions.
Kansas should mandate rural proofing for AI tools used in health and social service programs, starting with KDHE, KDADS, and DCF, supported by OITS. This involves agencies inventorying current and planned high-impact AI systems that affect eligibility, benefits, and care access, including vendor-provided tools. Before procuring or deploying these systems, a rural-proofing review must be conducted to ensure reliable performance in rural settings, sufficient rural data for validation, and that the system does not create undue burdens on areas with limited resources. Kansas should also establish a centralized, cross-agency AI governance approach to define statewide oversight, reporting, and minimum standards, requiring vendors to disclose performance in rural settings, data limitations, and human review points in plain language.
Kansas needs to move beyond simple consultation to co-governance, institutionalizing regular listening sessions with rural communities through trusted local partners. These sessions, coordinated by OITS or the Governorâs Office with research support from Kansas public universities, should involve KDHE, KDADS, and DCF. The goal is to surface needs, barriers, and unintended harms of AI in rural areas, especially given the magnified impact of design flaws in resource-constrained environments. Findings from these sessions should be integrated into state oversight processes to inform procurement, monitoring, and accountability for AI systems, strengthening the rural-proofing mechanism.
Kansas should create a publicly available âRural AI Health Governance Blueprintâ that details its rural AI governance framework, including rural-proofing standards, workflows, listening session models, and transparency practices. This framework, developed by OITS, ITEC, and participating agencies, should be regularly assessed and evaluated by Kansas public universities to gather evidence-based recommendations for continuous improvement. Kansas should then share these tools and model policies through interstate networks (e.g., NGA, NCSL) and pilot collaborations with other rural states to test the replicability of its workflow, positioning itself as a practical demonstration state for responsible rural AI governance that addresses resource, environmental, and land use impacts.
Kansas needs a statewide Rural Health AI Literacy Framework to ensure residents, students, and frontline workers can critically engage with AI systems, focusing on how AI influences health access, eligibility, and related decisions in rural communities. This framework, developed by KSDE, KBOR, and OITS, should span Kâ12, postsecondary education, and public-sector roles. Implementation includes integrating AI literacy into various curricula, embedding modules in higher education, establishing rural AI literacy hubs (e.g., Kansas State University and Cooperative Extension), expanding community programming through local partners, and providing baseline AI literacy training for state employees in health and human service roles.
As AI becomes increasingly integrated into public systems affecting health and social outcomes, it is crucial to consider rural contexts, especially in Kansas, where communities often operate with data sparsity, lower service density, and limited institutional oversight. The recommendations propose operationalizing responsible AI principles through coordinated cross-agency governance, integrating rural proofing into existing structures, and fostering stronger community engagement in AI decision-making. By taking these steps, Kansas can develop an accountable model for rural AI governance, providing a practical path forward for other rural states.