From Silos to Synergy: Artificial Intelligence as an Architect of Interoperable One Health Systems for Antimicrobial Resistance Surveillance ()
1. Introduction
Antimicrobial resistance has been characterized as a “silent pandemic” that fundamentally threatens the viability of modern medicine, undermining the efficacy of antibiotics that have served as the cornerstone of infectious disease management for nearly a century [1]. The World Health Organization estimates that bacterial resistance directly claims approximately 1.27 million lives annually, with an additional 4.95 million deaths associated with resistant infections [2]. Without transformative intervention, AMR could precipitate 10 million annual deaths by 2050, imposing an economic burden exceeding USD 100 trillion a catastrophe that would disproportionately devastate low- and middle-income countries where health systems are least resilient [3].
The One Health framework, which formally recognizes the inextricable linkages between human, animal, plant and environmental health, has been endorsed by the Quadripartite organizations (FAO, UNEP, WHO, WOAH) through the One Health Joint Plan of Action (2022-2026) as the guiding paradigm for coordinated AMR response [4]. This framework implicitly mandates integrated surveillance systems and seamless data sharing across sectors [5]. Yet nearly a decade after the adoption of the WHO Global Action Plan on AMR, a critical implementation gap persists: the information infrastructure required to operationalize One Health surveillance remains deeply fragmented, characterized by sector-specific data silos, incompatible standards and minimal interoperability [6] [7].
The consequences of this fragmentation are profound and multifaceted. Surveillance systems essential for tracking resistance patterns and guiding evidence-based interventions remain underdeveloped in many regions, constrained by inadequate infrastructure, limited laboratory capacity and the absence of standardized reporting protocols [8]. Multisectoral, multidisciplinary AMR research remains exceptional rather than normative and substantial knowledge gaps impede the development of integrated interventions [7]. Divergent analytical methodologies across sectors for instance, culture-based methods versus genomic approaches generate data that resist meaningful integration, partially explaining why One Health solutions have been slow to materialize despite robust conceptual endorsement [7].
This perspective paper advances a distinctive argument: the most pressing challenge confronting AMR mitigation is not a scarcity of data or analytical sophistication, but rather the fundamental incapacity of existing information systems to share and integrate data meaningfully across human, veterinary and environmental health domains [9] [10]. Artificial intelligence, when strategically oriented toward interoperability, offers a transformative pathway forward. However, realizing this potential demands a conceptual reorientation: we must move beyond AI’s celebrated applications in drug discovery and resistance prediction to focus on the foundational, less glamorous work of architecting systems that enable communication across sectoral divides [11] [12].
2. The Interoperability Imperative in One Health AMR Surveillance
Interoperability the capacity of diverse information systems to exchange and utilize data constitutes the sine qua non of effective One Health surveillance [13]. Yet achieving genuine interoperability across human, animal and environmental health sectors presents formidable challenges that are simultaneously technical, organizational and political [14].
At the technical level, data generated across sectors are characterized by profound heterogeneity: human clinical data typically reside in electronic health records structured around diagnostic and treatment codes, veterinary data are often scattered across livestock production systems and companion animal practices and environmental data derive from wastewater surveillance, agricultural runoff monitoring and ecosystem health assessments [15]. These data are collected using disparate instruments, stored in incompatible formats and governed by divergent quality standards [16].
Organizational barriers are equally consequential: health, agriculture and environmental agencies operate under separate mandates, funding streams and accountability structures that disincentivize cross-sectoral collaboration [17]. Professional cultures differ markedly, with human medicine oriented toward individual patient outcomes, veterinary medicine focused on herd health and food safety and environmental science emphasizing ecosystem integrity [18]. These cultural differences manifest in divergent data collection priorities, analytical approaches and communication practices that perpetuate sectoral silos [19].
Political and regulatory barriers compound these challenges. Data governance frameworks addressing ownership, privacy and usage rights are often absent or inconsistent across jurisdictions, creating legal uncertainty that discourages data sharing [20]. In many contexts, data protection regulations that rightfully safeguard patient privacy inadvertently impede the data aggregation necessary for robust surveillance [21].
The imperative for interoperability is not merely technical convenience but epidemiological necessity. AMR does not respect sectoral boundaries; resistant genes and organisms flow seamlessly between humans, animals and environmental reservoirs [22]. Effective surveillance must trace these pathways, identify transmission hotspots and evaluate intervention effectiveness across the entire transmission chain [23]. Without interoperable systems, this comprehensive picture remains elusive and interventions risk being misdirected or insufficient.
3. AI as an Interoperability Architect: Opportunities and Frontiers
Before turning to specific techniques, it is worth being precise about what “AI as an interoperability architect” means, since the phrase risks overstating what any algorithm can do. Interoperability comprises at least four distinguishable elements: i) data exchange, the technical transport of information between systems; ii) semantic integration, the alignment of meaning across differing vocabularies, ontologies and classification schemes; iii) governance, the political and legal arrangements that determine who may access, share and act on data; and iv) decision support, the translation of integrated data into outputs that clinicians, veterinarians and policymakers can trust and use. AI makes its most direct contribution to ii) and iv)—natural language processing can align heterogeneous vocabularies and explainable AI can render outputs interpretable. Its contribution to i) is more modest, chiefly through architectures such as federated learning that reduce the need to move raw data at all. Governance iii), by contrast, is fundamentally a sociotechnical and institutional prerequisite: no algorithm can substitute for the political agreements, legal frameworks and institutional trust that determine whether data-sharing is permitted in the first place. The sections that follow examine each AI technique against this framework, distinguishing what AI can realistically deliver from what remains contingent on human and institutional decisions.
Artificial intelligence offers a suite of techniques that can systematically address interoperability barriers. However, a critical insight often overlooked in AI-for-AMR discourse is that these techniques are not interchangeable; each addresses a distinct dimension of the interoperability challenge and their strategic combination is essential for comprehensive progress (Figure 1) [24].
Figure 1. AI as an interoperability architect across One Health sectors. Federated learning, natural language processing and explainable AI each address a distinct interoperability dimension, bridging human, animal and environmental data silos into integrated AMR surveillance.
3.1. Federated Learning: Privacy-Preserving Data Federation
Perhaps the most fundamental obstacle to integrated AMR surveillance is the legal, ethical and competitive constraints that prevent the sharing of sensitive data across institutional and sectoral boundaries [25]. Federated learning directly addresses this challenge by enabling AI model training across distributed datasets without requiring raw data movement, thereby substantially reducing, though not eliminating, as discussed below the privacy burden while enabling collaborative intelligence [26].
A cross-sectoral (human-animal-environmental) federated learning deployment specifically for AMR surveillance has not, to our knowledge, yet been documented in the peer-reviewed literature; readers should treat this as an unrealized but technically feasible opportunity rather than an established practice. Proof-of-concept evidence from clinical federated learning is nonetheless instructive: in a ten-institution consortium training a shared model without any raw patient data leaving participating sites, the resulting model reached 99% of the predictive quality achieved by a fully centralized model trained on pooled data [27]. Translating this architecture to link hospital, veterinary and environmental surveillance nodes for AMR would preserve data sovereignty, accommodate diverse regulatory frameworks and enable cross-sectoral collaboration that centralized data-sharing arrangements currently preclude provided the safeguards discussed below are in place.
This last caveat matters. Federated learning reduces, rather than eliminates, disclosure risk: shared model updates and gradients can still be vulnerable to gradient-inversion or membership-inference attacks capable of partially reconstructing source data [25]. Treating federation alone as a sufficient privacy guarantee would therefore be premature. Realizing its privacy benefits in practice requires complementary safeguards secure aggregation protocols that prevent any single party from inspecting individual updates, formal privacy-risk assessments conducted before deployment and clear institutional accountability for model governance across participating sites [26].
For low- and middle-income countries (LMICs), where data protection infrastructure may be nascent and data governance frameworks often underdeveloped, federated learning offers a technically feasible pathway to participation in global surveillance networks without compromising local data sovereignty [28]. This is particularly significant given that LMICs bear the greatest AMR burden yet often face the most stringent constraints on data sharing [29].
3.2. Natural Language Processing: Harmonizing Heterogeneous Data
The data that could inform AMR surveillance exist in an extraordinary variety of forms: structured laboratory results, free-text clinical narratives, veterinary diagnostic reports, environmental monitoring records, agricultural production data and policy documents [30]. Natural language processing (NLP) offers a powerful toolkit for extracting structured, machine-readable information from these heterogeneous sources, thereby enabling semantic interoperability the capacity to understand and integrate information across diverse data representations [31].
Recent advances in transformer-based language models have dramatically expanded NLP’s capabilities in the biomedical domain [32]. Researchers have successfully employed machine learning frameworks combining transformer architectures and cluster analysis to analyze over 44,000 scientific publications, identifying key AMR research topics, subtopics and detection methods across One Health sectors [33]. More ambitiously, large language models have been developed to assist AMR policy development by synthesizing evidence from national action plans across 146 countries [34]. The AMR-Policy GPT, a large language model with advanced retrieval-augmented generation capabilities, can search and summarize evidence from plans, metadata and technical knowledge, providing traceable references from global document databases [34].
It is important to separate what this evidence actually demonstrates. The topic-clustering application [33] and the policy-synthesis application (AMR-Policy GPT) [34] both show NLP performing well on text corpora scientific literature and policy documents, respectively that are relatively well-structured and written in standardized professional language. Neither constitutes evidence that NLP can reliably extract and harmonize information from the routine, often noisy clinical, veterinary and environmental surveillance records that this paper’s interoperability argument actually depends on. Peer-reviewed evaluations of NLP performance specifically on cross-sector routine-record harmonization remain limited; the methodological and ontological groundwork exists [30] [31], but validated performance evidence at the scale needed for operational surveillance is, at present, an evidence gap rather than a demonstrated capability.
The implications of these developments are profound. By converting unstructured text into standardized, machine-readable data, NLP enables the data integration essential for true One Health surveillance. However, critical challenges remain: NLP models must be trained on representative data, evaluated for bias and continuously validated to ensure accuracy across diverse linguistic and cultural contexts [35].
3.3. Explainable AI: Building Trust and Enabling Clinical Translation
Even the most accurate predictive models will fail to influence practice if clinicians, veterinarians and policymakers cannot understand or trust their outputs [36]. The escalating challenge of AMR demands clinician-ready machine learning models that are not merely accurate but interpretable [37]. By treating resistance genes as independent features and augmenting them with curated single-nucleotide polymorphisms and contextual markers, explainable AI approaches deliver scalable, transparent predictions aligned with clinical decision-making needs [38].
Transparency in AI predictions empower evidence-based prescribing decisions, reduces treatment failure and enables tailored care for patient subgroups at greatest risk of resistance [39]. When models remain interpretable, predictions align with the clinician’s mental model of resistance mechanisms, thereby bridging the gap between algorithmic output and clinical action [40]. In the high-stakes environment of AMR management, where treatment decisions have life-or-death consequences, explainable AI is not merely a technical preference but an ethical imperative [41].
Table 1 summarizes these three modalities, the interoperability dimension each addresses and their respective benefits and limitations.
Table 1. Comparison of AI techniques for interoperable One Health AMR surveillance.
AI technique |
Interoperability dimension addressed |
Core mechanism |
Illustrative application |
Key benefit |
Key limitation |
Federated learning |
Legal/organizational enables collaboration without moving raw data across institutional or sectoral boundaries |
Trains a shared model across distributed datasets held locally at each site |
Clinical proof-of-concept: 10-site federated model reached 99% of centralized-model quality; cross-sectoral AMR application not yet demonstrated |
Preserves data sovereignty and privacy; well suited to LMIC regulatory environments |
Requires reasonably harmonized data structures across sites before training can begin; cross-jurisdictional governance still unresolved |
Natural language processing |
Semantic converts unstructured, heterogeneous text into standardized, machine-readable data |
Extracts structured information from clinical narratives, veterinary reports and policy documents using transformer-based models |
AMR-Policy GPT synthesizing evidence from national action plans across 146 countries |
Unlocks the large volume of
AMR-relevant knowledge trapped in free text |
Accuracy depends on representativeness of training data; needs ongoing validation across languages and cultural contexts |
Explainable AI (XAI) |
Clinical/translational builds the interpretability needed for adoption by end users |
Produces transparent, feature-level rationale for predictions (e.g., resistance genes, SNPs) rather than opaque outputs |
Interpretable models aligning with clinicians’ mental models of resistance mechanisms |
Increases clinical trust and uptake; supports evidence-based prescribing |
Explainability can trade off against raw predictive performance; still an emerging methodological area |
4. Persistent Barriers to AI-Enabled Interoperability
Despite these promising opportunities, several significant challenges threaten to undermine AI’s contribution to interoperable One Health systems. These barriers are not merely technical but structural, requiring coordinated action at multiple levels [42].
4.1. Ontological Discordance and Data Standardization
AI models are only as robust as the data on which they are trained [43]. Yet across human, animal and environmental health sectors, data are collected using disparate methods, definitions and formats that resist harmonization. Divergent analytical methodologies such as culture-based versus genomic approaches generate data that cannot be meaningfully compared, partially explaining why One Health solutions have been slow to materialize [7].
The challenge extends beyond technical data formats to encompass fundamental ontological differences: how different sectors conceptualize and classify resistance, disease and health. For instance, the human health sector typically classifies resistance by clinical significance for patient outcomes, while the veterinary sector emphasizes implications for food production and animal welfare and the environmental sector focuses on ecological persistence and transmission dynamics [44]. These different conceptual frameworks generate data that are not merely stored differently but ask fundamentally different questions [45].
Greater harmonization of research methods, data standards and ontological frameworks across sectors is urgently needed [7]. Effective One Health information systems depend on standardized data protocols, interoperable platforms and strong data sharing networks [46]. Without common standards, ontologies and application programming interfaces, even the most sophisticated AI will struggle to integrate information meaningfully [47].
4.2. Infrastructural Deficits and Capacity Gaps
In LMICs, where the AMR burden is often greatest, limited digital access, funding shortages and lack of technical expertise severely constrain AI implementation [48]. Challenges related to user engagement and incentives, data quality and standardization, infrastructure and connectivity, capacity building and policy alignment are consistently identified as barriers to effective digital One Health surveillance [49]. This unevenness in equipment and technical capacity across countries particularly acute in the low- and middle-income settings where antimicrobial use and resistance are often highest means that any AI-enabled surveillance strategy must be phased and context-appropriate rather than assuming uniform readiness for deployment.
Even in countries with robust infrastructure, incompatible databases, siloed institutions and lack of common standards impede progress [50]. The challenges are compounded by the need for sustained capacity building, reliable electricity and internet connectivity that many regions lack [51]. Investment in capacity development in LMICs must be prioritized, as these are settings where the AMR burden is most acute and the benefits of effective surveillance greatest [49].
4.3. Regulatory and Governance Fragmentation
Data sharing across sectors raises complex questions of ownership, usage rights and privacy [52]. Clear governance frameworks defining these parameters are essential for confident AI deployment, yet such frameworks are often absent or inconsistent across jurisdictions [53]. The Quadripartite organizations have called for coordinated surveillance systems and launched a Joint Offer for targeted One Health investment, but translating this vision into operational reality requires national-level political commitment and investment that remains uneven [54].
The fragmentation of regulatory frameworks creates particular challenges for federated learning approaches, which operate across institutional and jurisdictional boundaries. Without clear guidance on data ownership, accountability and liability, institutions remain reluctant to participate in cross-sectoral AI initiatives [55].
5. Toward a Roadmap for Operationalizing AI-Enabled One Health Surveillance
The implications of this analysis demand urgent attention from policymakers, researchers and funders [56]. We propose a pragmatic roadmap comprising four interconnected priorities.
5.1. Prioritize Interoperability Infrastructure over Isolated Predictive Tools
Investment in AI for AMR must prioritize foundational interoperability infrastructure over isolated predictive tools [57]. This means dedicating resources to common data standards, shared platforms, governance frameworks and capacity building—not merely algorithmic innovation [58]. When embedded within well-regulated One Health frameworks, AI can strengthen surveillance, improve treatment decisions and support evidence-based interventions [59]. However, this requires deliberate investment in the foundational architectures that enable data to flow across sectors [60].
5.2. Elevate Explainability as a Non-Negotiable Requirement
Explainability must be treated as a non-negotiable requirement for clinical AI deployment [61]. Transparency empowers evidence-based prescribing decisions and reduces treatment failure [38]. AI should be positioned as a decision-support tool that augments, rather than supplants, clinical reasoning [62]. By keeping models interpretable, predictions align with clinical mental models of resistance mechanisms, enabling trust and adoption [40]. This is not merely a technical preference but an ethical and practical imperative for effective AMR management [41].
5.3. Develop Context-Appropriate Solutions
Solutions must be context-appropriate rather than imposing high-resource models on LMIC settings [63]. Offline-capable mobile applications, low-bandwidth platforms and locally relevant training programs may be more suitable than sophisticated cloud-based systems [64]. Co-design with local stakeholders, infrastructure investment, applications integration, cross-sectoral collaboration, continuous training, evidence-based policy support and sustainable funding models are crucial for building effective, inclusive and scalable ICT-driven solutions [65].
5.4. Leverage Current Political Momentum
The current political momentum provides an unprecedented opportunity [66]. The Quadripartite Joint Offer for targeted One Health investment and the inclusion of AMR in the WHO Pandemic Agreement negotiations signal growing political will [67]. However, opportunity alone is insufficient. Realizing AI’s potential to combat AMR requires deliberate, coordinated and adequately funded effort to build the interoperable information systems that One Health demands [67].
5.5. Establish an Operational Implementation Sequence
The priorities above remain aspirational unless translated into an actionable sequence. We propose the following operational steps, each a precondition for the next: 1) agree a minimum set of shared data elements that every participating sector commits to reporting in a common format; 2) map these elements onto existing standards and ontologies, resolving the definitional discordance identified in Section 4.1 before any data are pooled or federated; 3) validate data quality at each participating site against agreed thresholds, since AI trained on unvalidated data will simply propagate existing inconsistencies; 4) secure governance and ethics approval covering data ownership, accountability and liability across all participating institutions and jurisdictions; 5) pilot a federated deployment at limited scale, incorporating the privacy safeguards discussed in Section 3.1; and 6) scale deployment only after pilot results are independently evaluated, with continuous monitoring for model drift and equity of access. Sequencing matters: attempting model deployment before data-quality validation and governance approval are in place has been a recurring failure mode in digital One Health initiatives [49] [50].
6. Conclusion
The time for siloed, sector-specific approaches to AMR surveillance has passed. The question is no longer whether AI can help combat AMR, but whether we have the collective will to deploy it in service of the integrated surveillance systems that the crisis demands. Federated learning, natural language processing and explainable AI offer promising pathways to overcome interoperability barriers, but they require intentional investment, context-appropriate implementation and robust governance frameworks. As the global community mobilizes to address AMR through the One Health framework, we must ensure that interoperability is not an afterthought but a foundational principle. AI can help us build bridges between silos, but only if we have the vision and commitment to build them together. We intend this perspective for policymakers and funders as much as for technical audiences, since the barriers described here are institutional and political at least as much as they are algorithmic. This perspective calls for a fundamental reorientation: interoperability must be elevated from a technical afterthought to a central pillar of global AMR strategy, with AI as the enabling architecture that makes integrated One Health surveillance possible.
Acknowledgements
The authors extend sincere gratitude to Professor Raywat Deonandan for invaluable intellectual guidance, scholarly mentorship and critical oversight throughout the development of this perspective paper.
Author Contributions
Ally Mwambela conceived and designed the study and drafted the manuscript. Said Hassan, Yuda Sule and Joel Makyao contributed to data acquisition, analysis and interpretation and critically reviewed the manuscript. Professor Raywat Deonandan provided critical revisions for important intellectual content and supervision. All authors reviewed and approved the final version and agree to be accountable for all aspects of the work.