AI for Public Health Emergency Response

AI for Public Health Emergency Response is sponsored by National Institutes of Health (NIH). Encourages AI research to improve public health responses during emergencies.

Use of Artificial Intelligence in Public Health Education for Pandemic Preparedness and Response - PMC As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. .

2026 Feb 20;92(1):21. doi: 10. 5334/aogh.

5130 Use of Artificial Intelligence in Public Health Education for Pandemic Preparedness and Response Ellen Crystian Silvestre Garcia Souza Ellen Crystian Silvestre Garcia Souza 1 Universidade Federal de Mato Grosso do Sul (UFMS), Três Lagoas Campus, MS, 79600-080, Brazil Find articles by Ellen Crystian Silvestre Garcia Souza 1 , Aires Garcia dos Santos Junior Aires Garcia dos Santos Junior 1 Universidade Federal de Mato Grosso do Sul (UFMS), Três Lagoas Campus, MS, 79600-080, Brazil Find articles by Aires Garcia dos Santos Junior 2 School of Nursing, Universidade de São Paulo, São Paulo, SP, 05403-000, Brazil Find articles by Adriana M S Félix 2 , João Paulo Assunção Borges João Paulo Assunção Borges 3 Universidade Federal Mato Grosso do Sul (UFMS), Coxim Campus, MS, 79400-000, Brazil Find articles by João Paulo Assunção Borges 3 , Layze Braz de Oliveira 4 Universidade Federal de Maran˙hão, São Luís 65080-805 Maranhao, Brazil Find articles by Layze Braz de Oliveira 4 , Liliane Moretti Carneiro 1 Universidade Federal de Mato Grosso do Sul (UFMS), Três Lagoas Campus, MS, 79600-080, Brazil Find articles by Liliane Moretti Carneiro 1 , Alvaro Francisco Lopes de Sousa Alvaro Francisco Lopes de Sousa 1 Universidade Federal de Mato Grosso do Sul (UFMS), Três Lagoas Campus, MS, 79600-080, Brazil Find articles by Alvaro Francisco Lopes de Sousa 1 Universidade Federal de Mato Grosso do Sul (UFMS), Três Lagoas Campus, MS, 79600-080, Brazil 2 School of Nursing, Universidade de São Paulo, São Paulo, SP, 05403-000, Brazil 3 Universidade Federal Mato Grosso do Sul (UFMS), Coxim Campus, MS, 79400-000, Brazil 4 Universidade Federal de Maran˙hão, São Luís 65080-805 Maranhao, Brazil Received 2025 Dec 12; Accepted 2026 Jan 18; Collection date 2026.

This is an open-access article distributed under the terms of the Creative Commons Attribution 4. 0 International License (CC-BY 4. 0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

See https://creativecommons. org/licenses/by/4. 0/ .

PMCID: PMC12927464 PMID: 41736834 Background: The rapid evolution of artificial intelligence (AI) has enabled new approaches for health education, particularly during public health emergencies. However, evidence remains fragmented on how AI-based educational strategies support preparedness, response, and recovery phases of pandemics and epidemics.

Objective: To map the use of AI-based technologies in health education strategies addressing preparedness, response, and recovery during public health emergencies, identifying target populations, intervention characteristics, outcomes, scalability, and knowledge gaps. Methods: This scoping review followed Joanna Briggs Institute methodology and PRISMA-ScR guidelines.

Searches were conducted in PubMed/MEDLINE, Scopus, Web of Science, Embase, IEEE Xplore, and LILACS, complemented by gray literature from Google Scholar. Studies published from 2010 onward in English, Portuguese, or Spanish were included. Eligible designs comprised primary studies, methodological or implementation research, and reviews with explicit educational components.

Data extraction covered context, populations, AI modalities, educational purposes, delivery channels, supervision requirements, pandemic-cycle phase, scalability, outcomes, and evidence gaps. Results: Forty-one studies met the inclusion criteria. Conversational AI (chatbots and large language models) and algorithmic curation tools using machine learning and natural language processing predominated.

Most interventions supported health literacy, risk communication, and misinformation management; others addressed personalized learning, microtraining, and clinical simulation for students and health professionals. Delivery channels included mobile applications, messaging platforms, websites/YouTube, and clinical AI systems.

Human oversight (expert validation and curation) was consistently reported as essential for safety and reliability. Interventions mainly targeted the response phase, with emerging applications for preparedness. Major gaps included standardized learning measures, cost-effectiveness evaluations, equity analyses, and governance frameworks ensuring privacy, transparency, and bias control.

Conclusions: AI-enabled educational technologies can strengthen rapid, scalable, and personalized learning during health emergencies. Future research should prioritize multicenter studies using standardized indicators, economic and equity assessments, and robust governance frameworks to ensure ethical, safe, and inclusive adoption.

Keywords: artificial intelligence, public health education, pandemics, chatbots, machine learning, misinformation Public health emergencies experienced in recent decades, including the COVID-19 pandemic and the monkeypox outbreak, as well as earlier events such as SARS (2003), H1N1 influenza, Ebola, Zika, and the reemergence of poliomyelitis, all classified by the World Health Organization (WHO) as Public Health Emergencies of International Concern (PHEICs), have forcefully exposed the vulnerability of health systems [ 1 ].

These critical scenarios have brought to light major structural and operational weaknesses in health systems, especially in epidemiological surveillance and in their capacity to prepare for, respond to, and recover efficiently from health crises.

In addition, they have revealed gaps in professional training, health communication, and institutional agility, underscoring the urgency of more resilient, intersectoral, and integrated systems [ 1 ].

In this context, public health education plays a strategic role by strengthening the capacity for early risk identification, situational analysis, and decision-making in the face of emergencies, integrating technical, scientific, and technological competencies focused on prevention, harm mitigation, surveillance, and rapid response [ 1 ].

Among the innovations driving this transformation, artificial intelligence (AI) stands out, with applications ranging from epidemiological modeling and outbreak forecasting to clinical decision support and the personalization of educational processes [ 1 , 2 ].

During the COVID-19 pandemic, AI-based solutions, such as adaptive platforms, intelligent tutors, chatbots, and recommender systems, made it possible to maintain professional training while expanding access to health information, even in the context of mobility restrictions and physical distancing [ 1 , 2 ].

This integration between AI and education has proven essential for preparing professionals to work in complex scenarios characterized by high uncertainty, large data volumes, and the need for rapid, evidence-based decisions [ 3 ].

Consistent with this perspective, the WHO emphasized in a 2018 report that digital technologies and AI are central instruments for achieving global health goals, such as expanding universal coverage, protecting populations from emergencies, and promoting the well-being of billions of people [ 4 ].

In emergency management, AI has helped improve real-time monitoring, strengthen intersectoral coordination, optimize resources, and build the capacity of response teams [ 5 ]. At the same time, intelligent systems have expanded post-pandemic surveillance capacity, enabling early threat detection and the identification of emerging epidemiological patterns [ 6 ].

Despite this transformative potential, ethical and structural challenges remain, related to algorithmic bias, data privacy, and inequalities in access to digital infrastructure [ 1 , 3 ]. Given this scenario, integrating AI into educational strategies constitutes an emerging and promising frontier in public health, with the potential to enhance professional training and performance during health crises.

However, syntheses that consolidate the evidence on this interface are still scarce. Thus, this scoping review aims to map the evidence on the use of AI-based technologies in educational strategies targeting the preparedness, response, and recovery phases of public health emergencies, including pandemics, epidemics, and other events, by describing interventions, target populations, technologies, outcomes, and knowledge gaps.

This research was conducted as a scoping review, following the methodological framework proposed by Arksey and O’Malley and further developed by the Joanna Briggs Institute (JBI) [ 7 ].

The protocol was structured to ensure transparency and reproducibility of the process, following the steps: (1) identification of the research question; (2) identification of relevant studies; (3) study selection; (4) data extraction; and (5) collection, synthesis, and reporting of results.

Accordingly, the research question and key search elements for this review were developed using the PCC strategy, a mnemonic that helps identify the core topics: problem, concept, and context.

In this review, the problem was defined as the use of AI-based technologies; the central concept was public health educational strategies; and the context was the preparedness, response, and recovery phases in the setting of pandemics, epidemics, and other emergencies.

Thus, the guiding research question was: What scientific evidence exists on the use of AI-based technologies in educational strategies targeting the preparedness, response, and recovery phases of public health emergencies, including pandemics, epidemics, and other events of public health relevance?

Searches were performed in major bibliographic databases, PubMed/MEDLINE, Scopus, Web of Science, Embase, IEEE Xplore, and LILACS, and complemented by a targeted search of gray literature in Google Scholar, limited to the first pages ranked by relevance, as well as institutional documents cited in the included studies, such as those from the World Health Organization (WHO), the Pan American Health Organization (PAHO), and the Centers for Disease Control and Prevention (CDC).

The strategies combined controlled descriptors and free-text terms, tailored to each database, with Boolean operators, truncation, and restriction to title, abstract, and subject fields. Filters were applied for the period from 2010 onward and for publications in English, Portuguese, and Spanish. Complete search strategies for each database were archived to support reproducibility, and searches were last updated in October 2025.

Eligible studies included primary quantitative, qualitative, or mixed-methods research, as well as methodological studies and implementation reports that described the explicit use of AI for educational purposes in public health in either real or simulated pandemic scenarios. Narrative and integrative reviews and bibliometric studies were also included when they provided syntheses applicable to educational practice.

We excluded reports without an educational component (for example, purely technical surveillance or diagnostic applications), editorials and opinion pieces without methodological or implementation contributions, studies without full-text access, and publications in languages other than English, Portuguese, and Spanish.

Record management included export in RIS and BibTeX formats and integration into a bibliographic manager, followed by two-step deduplication: an automated software process (based on title, authors, year, and DOI) and manual inspection for residual cases.

Study selection was conducted in two phases by two independent reviewers: title and abstract screening based on eligibility criteria and full-text assessment of potentially relevant records. Disagreements were resolved by consensus and, when necessary, by a third reviewer. Records without full-text access, after attempts to contact authors or institutions, were excluded with appropriate justification.

The eligibility flow was documented in a PRISMA-ScR diagram.

Data extraction was performed using a pilot-tested form that captured: identification (author, year, country), public health emergency addressed, target population and channel/setting, AI modality and educational purpose, platform and presence of human curation/supervision, phase of the pandemic cycle, operational and scalability aspects, study type, and main reported limitations.

When available, we also recorded educational outcomes, engagement and acceptability metrics, references to equity and language, costs/implementation, and ethical considerations. Missing information was explicitly classified as “not reported (NR),” avoiding unsupported imputations.

The synthesis combined descriptive and narrative approaches, organized into three complementary tables: (i) characterization of studies (identification, context, population, and design); (ii) AI interventions and strategies with their educational objectives, channels, and supervision; and (iii) functional synthesis linking the educational role of each intervention to pandemic cycle phases, AI-mediated pedagogical mechanisms, scalability/operability conditions, level of evidence, and key gaps.

The interpretive narrative integrated convergent findings and contextual variations, as well as recurrent absences of standardized metrics for educational and behavioral effects, economic evaluations, and equity analyses. Given the inherent scope of mapping reviews, we did not conduct a formal risk-of-bias assessment.

To strengthen interpretability, we explicitly reported the study type and limitations declared by the authors (for example, restricted samples, dependence on specific platforms, language bias, and absence of clinical outcomes) and discussed their impact on the generalizability and applicability of the findings.

This strategy supports methodological transparency and coherence between the aim of mapping the field and the level of inference that is appropriate based on the available body of evidence. Table 1 presents a synthesis of 31 studies published between 2020 and 2025, encompassing different methodological designs, ranging from narrative and integrative reviews to observational and experimental studies.

A predominance of exploratory research and narrative reviews was observed, highlighting the still-emerging and consolidating field of AI in educational strategies applied to public health.

Most studies were conducted in the context of the COVID-19 pandemic, although some initiatives also addressed other epidemic situations or post-pandemic phases, with an emphasis on surveillance, risk communication, and training of health professionals. Methodological and contextual characteristics of the studies included in the review ( n = 31), Brazil, 2020–2025.

Methodological study (ML) Algorithmic curation of videos (YouTube); proposal of chatbots and integration into apps Public/professionals; YouTube/apps Improves the quality and discovery of trustworthy content; feasible integration with official channels No behavioral evaluation in real-world settings; platform dependence AI chatbots for education/clinical tutoring Students/educators; digital environment Potential for personalized tutoring and post-pandemic support Heterogeneous evidence; no clinical outcomes Franchini et al.

(2021) [ 10 ] Implementation study (mixed methods) Community chatbot (Dress-COV) for triage/education Reach and interaction with participatory education Nonequivalent control; limited generalizability Văduva et al.

(2023) [ 11 ] eHealth/mHealth/telemedicine (includes AI) Expands access and remote training Non-systematic; no effect metrics Digital transformation (includes AI) Emphasizes the educational role of digital health Abdelouahed et al.

(2025) [ 1 ] Exploratory qualitative study Adaptive AI; simulators; personalized content Continuous training and tailored materials Documentary/case-based; no measurement Public questions (COVID-19) Accessible and rapid responses Qualitative assessment; LLM biases Sezgin and Kocaballi (2025) [ 14 ] Generative AI in messaging (WhatsApp/SMS) Frequently asked public health questions Greater clarity and accuracy of responses McKee et al.

(2025) [ 15 ] Data/AI for segmented communication Public health professionals Popular and digital education with greater impact Haupt et al. (2024) [ 16 ] Experimental study (prompts) Media literacy/AI (role-playing game versus neutral) Better misinformation detection with appropriate prompting Tanui et al. (2024) [ 17 ] Apps with multilingual AI African populations (general) Inclusive and scalable education Bharel et al.

(2024) [ 18 ] Generative AI for communication/efficiency Health professionals/organizations Reduces administrative burden; supports messaging ChatGPT (health questions) Good performance on educational FAQs Zeeb et al. (2023) [ 20 ] Awareness via apps during COVID-19 Towler et al.

(2023) [ 21 ] ML (topic modeling) for rapid analysis of qualitative data Accelerates insights for communication Does not measure public impact Digital health curriculum with AI Need for curricular integration and practice Self-reported; non-experimental Training in surveillance with AI Training plus real-time alerts No educational measurement AI in diagnosis/CT (with educational pathway) Training for clinical AI use Clinical focus; indirect education Grüne et al.

(2022) [ 25 ] Retrospective observational study Symptom app with feedback Use bias; no counterfactual Weeks et al. (2022) [ 26 ] Personalized chatbot for vaccine hesitancy Empathic messages increase acceptance Qualitative; no population-level effect Dzau et al.

(2022) [ 27 ] Digital capacity-building frameworks Proposes simulations and remote teaching Narrative review/conceptual paper AI-enhanced curriculum; data-driven teaching Public health students/educators; university courses Framework to integrate AI and big data into public health education Conceptual paper; no empirical evaluation Trends in digital/AI research Narrative review/perspective Adaptive learning; AI tutoring; simulations Public health/medical students and professionals; Digital platforms/simulation-based training AI can personalize learning and support simulation-based public health training at scale Theoretical overview; no primary data or implementation studies Scott and Coiera (2020) [ 29 ] Critical narrative review Early warning/NLP and modeling Supports policies and messaging No direct educational assessment Uohara et al.

(2020) [ 30 ] Triage chatbots; telemonitoring Scales recommendations and recruitment Montenegro-López (2020) [ 31 ] National app plus AI committee Guidance and local management Simsek and Kantarci (2020) [ 32 ] Optimized allocation (AI) Informs planning/education No direct educational channel McKillop et al.

(2021) [ 33 ] Mixed-methods exploratory study COVID-19 chatbots based on CDC/WHO Positive use and acceptability Uncertain behavioral effect Verma et al. (2025) [ 34 ] Feasibility study (mixed methods) Hospital educational technology Improved compliance during the intervention Single-center; short term Bynon Neely et al.

(2024) [ 35 ] YouTube plus SEO with ChatGPT support Communities and health workers Engagement and reach of videos With respect to target audiences, the studies covered four broad groups: health professionals and managers; students and educators; communities and citizens; and specific populations, such as vaccine-hesitant youth and public health workers.

Educational approaches ranged from simulated training and remote teaching, aimed at developing digital competencies, to automated and interactive messaging designed to promote awareness and adherence to preventive measures.

Taken together, the studies highlight the potential of AI to optimize learning, personalize content, improve access to high-quality information, and expand the reach of educational initiatives, especially in contexts with restrictions on in-person activities.

A strong integration between AI and digital health was also observed, with applications that extend beyond educational environments to emergency management, participatory surveillance, and community engagement.

On the other hand, the authors point to recurring limitations, such as methodological heterogeneity, absence of educational impact metrics, platform bias, sample constraints, and dependence on specific technological infrastructure. These factors still hinder the comparability and generalizability of findings across the different contexts and studies analyzed.

Table 2 illustrates the breadth of AI-mediated educational strategies applied to the preparedness, response, and recovery phases of public health emergencies, reflecting significant advances in the integration of digital technologies with innovative educational practices.

A total of 31 studies were identified that explored different AI modalities, with particular emphasis on chatbots, adaptive learning platforms, ML models, natural language processing (NLP), and generative systems.

These technologies were used in multiple contexts and for diverse purposes, ranging from the curation and recommendation of trustworthy content to support for clinical training, participatory education, and health communication. AI-based educational strategies: Modality, purpose, platform, and curation ( n = 31), Brazil, 2025.

EDUCATIONAL PURPOSE (ESSENCE) ML + NLP for video curation Filter and recommend trustworthy videos to strengthen health literacy and reduce misinformation Multilingual potential; integration with official channels AI chatbots (integrative review) Personalized clinical tutoring/learning in the post-pandemic period Franchini et al.

(2021) [ 10 ] Community chatbot (Dress-COV) Triage plus participatory education and reinforcement of self-care Accessible; community inclusion Văduva et al. (2023) [ 11 ] eHealth/mHealth/telehealth (with AI) Remote training and adoption of digital technologies Editorial (AI in digital health) Emphasizes informing/training for safe use of technologies Abdelouahed et al.

(2025) [ 1 ] Adaptive AI; intelligent simulators Continuous training and profile-based personalized content Answer public questions in plain language Sezgin and Kocaballi (2025) [ 14 ] Generative AI in messaging Educational support; assess clarity and relevance of responses McKee et al. (2025) [ 15 ] Data + AI (applied review) Segmented communication and decision support in public health Haupt et al.

(2024) [ 16 ] Prompting (role-playing game) in LLM Media literacy and misinformation detection Tanui et al. (2024) [ 17 ] Apps with multilingual AI Inclusive, scalable education in African public health settings Bharel et al. (2024) [ 18 ] Generative AI (perspective) Support communication, productivity, and insights ChatGPT (performance evaluation) Complementary study/FAQ tool Zeeb et al.

(2023) [ 20 ] Towler et al. (2023) [ 21 ] Accelerate insights to guide campaigns Text data analysis environments Digital health curriculum (with AI) Curricular integration and simulated practice Distance/hybrid education AI in surveillance (review) Train professionals and issue real-time alerts Educational track for clinical AI use Grüne et al. (2022) [ 25 ] Real-time educational feedback and self-care Weeks et al.

(2022) [ 26 ] Personalized vaccine chatbot Empathic messages to reduce hesitancy Dzau et al.

(2022) [ 27 ] Simulations and continuing education Big-data AI; intelligent tutoring; virtual simulation Integrate AI into public health curriculum and build AI-literate, emergency-ready professionals University public health courses; computer-assisted and online learning Teacher-led; faculty control of AI tools No explicit equity or multilingual strategy mentioned Bibliometrics (AI/digital) Map trends to guide education/management Wang and Li, (2024) [ 3 ] Personalized learning algorithms; predictive analytics; AI-driven simulations Personalize public health training and support data-informed decision-making Digital learning platforms; simulation/VR; AI-enhanced online courses Educator/institutional oversight; emphasis on ethical governance Discusses fairness and bias; no concrete language/localization plan Scott and Coiera (2020) [ 29 ] NLP/early warning; modeling Support messaging and rapid response Uohara et al.

(2020) [ 30 ] Triage chatbots; ML for research Scaled recommendations and recruitment Web/telehealth/virtual ICU Montenegro-López (2020) [ 31 ] National app + AI committee Guidance and local management with user feedback Simsek and Kantarci (2020) [ 32 ] SOFM (optimized mobilization) Inform logistical planning/education Models/decision-support tools McKillop et al.

(2021) [ 33 ] Watson Assistant (chatbots) COVID-19 information based on CDC/WHO Watson Assistant chatbots Verma et al. (2025) [ 34 ] Education/compliance with NPIs in hospital environments CCTV + IEC campaigns (information, education, communication) Bynon Neely et al.

(2024) [ 35 ] ChatGPT for educational SEO Expand reach/discovery of health videos ML + NLP (detailed pipeline) Preselect relevant and comprehensible videos Among the tools analyzed, chatbots and conversational assistants stood out as the most frequently used, appearing in roughly half of the studies.

Applications such as Dress-COV, Watson Assistant, and personalized vaccine chatbots were widely employed for interactive education, automated triage, and reduction of vaccine hesitancy, showing positive results in terms of accessibility, communicative empathy, and adherence to preventive measures. In addition, platforms based on ML and NLP, such as those proposed by Guo et al. [ 8 ] and Towler et al.

[ 21 ], were applied to the curation of educational videos and automated topic analysis, contributing to the dissemination of high-quality information and to combating health misinformation. Adaptive learning systems and intelligent simulators demonstrated strong potential for personalized learning, tailoring content to users’ profiles and individual needs and promoting autonomy, engagement, and educational continuity.

These solutions were predominantly implemented in formal learning environments, such as universities and distance education platforms, generally under instructional or technical supervision. AI models integrated into telemedicine and mobile health (mHealth) further expanded opportunities for remote education and home-based support, particularly in low-connectivity settings, helping to reduce digital inequalities.

With regard to channels and platforms, instant messaging services (such as Telegram, WhatsApp, and SMS), mobile applications, web platforms, and hybrid learning environments predominated, underscoring the versatility of AI for large-scale communication and capacity building.

Although not all studies reported direct human supervision, expert- or technical committee-led content curation was identified as an essential good practice to ensure accuracy, reliability, and ethical standards in the dissemination of information.

Table 3 synthesizes the distribution of educational functions mediated by AI across the preparedness, response, and recovery phases of public health emergencies, highlighting the versatility and reach of these technologies in different educational and operational contexts.

Among the 30 studies analyzed, applications were predominantly concentrated in the preparedness (36%) and response (52%) phases, with fewer initiatives focused on recovery (12%), a trend that reflects the global emphasis on readiness, mitigation, and immediate response during critical periods. Educational functions by phase of the pandemic cycle, AI mechanisms, scalability, and gaps ( n = 31), Brazil, 2025.

PHASE (PREP/RESPONSE/RECOVERY) Curation of trustworthy content (literacy) General public/professionals ML/NLP for selection; multimedia delivery High: integrates with official channels Methodological (development + evaluation) Assess behavioral effect; platform dependence Tutoring/educational support post-pandemic Tutor chatbot (personalization) Heterogeneity; no clinical outcomes Franchini et al.

(2021) [ 10 ] Triage plus participatory community education Validated messages plus reinforcement Implementation study (mixed methods) Nonequivalent control group Văduva et al. (2023) [ 11 ] Digital capacity building for nurses eHealth/mHealth/telehealth with AI support Variable: depends on infrastructure Small sample; no effect measurement Abdelouahed et al.

(2025) [ 1 ] Continuous training and adaptive content Desirable (faculty/preceptors) Exploratory qualitative study Public FAQ in plain language Model bias; update issues Sezgin and Kocaballi (2025) [ 14 ] Clarity and relevance of conversational responses Generative AI in messaging High: ubiquitous channels McKee et al. (2025) [ 15 ] Data-driven segmented communication Public health professionals Haupt et al.

(2024) [ 16 ] Media literacy (misinformation detection) Role-playing game prompting in LLM Tanui et al. (2024) [ 17 ] Inclusive multilingual education Apps with AI (local languages) Descriptive evidence only Bharel et al. (2024) [ 18 ] Institutional communication and productivity Public health agencies/professionals Generative AI for summarization/generation Moderate: requires governance Zeeb et al.

(2023) [ 20 ] Awareness through regional apps Towler et al.

(2023) [ 21 ] Rapid insights from qualitative data ML (topic modeling) for rapid synthesis High: accelerates decision-making Integration of digital health/AI into curricula Distance/hybrid learning with AI Self-reported data; no impact outcomes Surveillance training and alerts AI for detection and alerts Training for clinical AI use (imaging) Moderate: requires infrastructure Retrospective data; variability Grüne et al.

(2022) [ 25 ] Self-care via symptom feedback Retrospective observational study Self-report; external validation Reduction of vaccine hesitancy Personalized chatbot (empathic messages) Dzau et al.

(2022) [ 27 ] Digital training and simulations Simulations and remote teaching Modernize public health curriculum and train emergency-ready professionals Public health students; public health educators AI-based intelligent tutoring; data-driven curriculum design; computer-assisted learning using big data University courses; online/computer-assisted Potentially high via e-learning; only proposed No implementation; no learning outcomes Mapping thematic frontiers Bibliometrics of digital/AI Coverage and language bias Personalize and modernize public health training Public health/medical students; health professionals Adaptive learning; AI simulations; analytics Digital platforms; online courses; simulation High theoretical scalability; not tested No primary data; few concrete models for LMICs Scott and Coiera (2020) [ 29 ] Data-informed messaging and rapid response NLP/early warning; modeling Critical narrative review No educational evaluation Uohara et al.

(2020) [ 30 ] Scaled recommendations and recruitment Triage chatbots; ML for research Web/telehealth/virtual ICU Montenegro-López (2020) [ 31 ] Guidance and local management National app plus AI committee Validation of decision rules; asymptomatic cases Simsek and Kantarci (2020) [ 32 ] Mobilization planning and logistics Models/decision-support tools Dependence on assumptions McKillop et al.

(2021) [ 33 ] Automated informational service Chatbots (Watson Assistant) Mixed-methods exploratory study No metrics for satisfaction, time, or cost Verma et al. (2025) [ 34 ] Compliance with NPIs in hospitals Video detection (YOLO-V5 + 3D) Moderate: hardware-dependent Feasibility study (mixed methods) No control group; confounding factors Bynon Neely et al.

(2024) [ 35 ] Multimedia educational reach Communities/public health workers History of misinformation Notes: Phases—preparedness (prep), response (response), and recovery (recovery). NR = not reported. “Level of evidence” refers to the type of study/report.

In the preparedness phase, AI was widely used in professional training processes and in the integration of digital health into curricula through intelligent simulations, adaptive learning systems, and instructional platforms. These approaches supported the development of technical and digital competencies, preparing professionals and students to work in scenarios characterized by risk, uncertainty, and information overload.

In addition, solutions based on modeling and predictive analytics were applied to optimize resource management, logistical planning, and institutional communication. In the response phase, most of the experiences described centered particularly on the use of chatbots, NLP systems, ML models, and multimedia platforms.

Tools such as Watson Assistant, Dress-COV, and national applications like CoronApp were widely employed for participatory education, automated triage, reduction of vaccine hesitancy, and media literacy, standing out for their high scalability and low operational cost.

In surveillance and risk communication contexts, AI was also applied to the curation of trustworthy content, monitoring of misinformation, and provision of personalized conversational responses, expanding the reach of educational messages and strengthening community engagement.

In the recovery phase, studies focused on initiatives for digital capacity building and psychosocial-educational support in the post-pandemic period, with emphasis on conversational tutoring models, eHealth/mHealth, and AI-assisted distance education aimed at professional requalification and the resumption of academic activities.

These experiences demonstrated the potential of AI to ensure educational continuity and reduce inequalities in access, although they still rely heavily on technological infrastructure and regional connectivity.

With regard to scalability, more than 70% of the experiences analyzed showed operational feasibility at scale, mainly through web platforms and mobile messaging services, which enabled broad dissemination of information and dynamic interaction with diverse audiences.

However, important gaps remain, such as methodological heterogeneity, the absence of educational impact metrics, dependence on commercial platforms, and the lack of behavioral and clinical indicators that would allow assessment of the actual effects on learning and practice change. Figure 1 presents a conceptual model that illustrates the applications of AI in public health education across the pandemic preparedness and response cycle.

The figure shows that AI technologies, especially chatbots, generative models, and ML systems, have been used predominantly in the response phase of health emergencies. These tools are delivered through mobile applications, web platforms, and messaging services (such as WhatsApp and Telegram), enabling broad dissemination of educational content.

Their main educational functions include personalization of learning, content adaptation, risk communication, reduction of misinformation, and rapid data analysis. Together, these AI-mediated strategies contribute to improving health literacy, increasing self-care, and promoting greater adherence to preventive measures among the population. AI framework for public health education and emergency response.

Role of artificial intelligence in public health education across pandemic phases This study mapped evidence on the use of AI-based technologies in educational strategies aimed at planning the preparedness, response, and recovery stages of public health emergencies.

The results of the studies assessed in this review reveal the consolidation of AI as a health education tool in the context examined, enabling risk identification, optimization of responses, and personalization of educational content for health professionals and the general public.

These findings are corroborated by a previous study that explored lessons learned during the COVID-19 pandemic and offered insights into the use of AI as support in crisis scenarios [ 1 ]. Public health emergencies constitute serious threats to population health, as they can result in widespread dissemination of infectious agents, high morbidity and mortality, and substantial impacts on the reorganization of health services [ 36 ].

In this context, AI emerges as a strategic technology by supporting rapid, data-driven decision-making, enhancing risk identification, and guiding timely interventions in epidemic and pandemic events.

However, despite its considerable potential, the effectiveness of AI solutions depends on structural factors, such as global collaboration, robust governance, compliance with ethical principles, standardization and validation of information, interoperability between systems, and assurance of equity in access to and use of technologies [ 1 , 37 , 38 ].

The literature indicates that AI has substantial potential as an educational resource and as support for actions across all stages of pandemic and epidemic management. In the preparedness phase, algorithms applied to the analysis of large datasets can be used to forecast outbreaks, inform public policy formulation, and support vaccine