AI-enhanced crowdsourcing for disaster management: strengthening community resilience through social media

Abstract

As disasters become more frequent and complex, the integration of artificial intelligence (AI) with crowdsourced data from social media is emerging as a powerful approach to enhance disaster management and community resilience. This study investigates the potential of AI-enhanced crowdsourcing to improve emergency preparedness and response. A systematic review was conducted using both qualitative and quantitative methodologies, guided by the PRISMA framework, to identify and evaluate relevant literature. The findings reveal that AI systems can effectively process real-time social media data to deliver timely alerts, coordinate emergency actions, and engage communities. Key themes explored include the effectiveness of community participation, AI’s capacity to manage large-scale information flows, and the challenges posed by misinformation, data privacy, and infrastructural limitations. The results suggest that when strategically implemented, AI-enhanced crowdsourcing can play a critical role in building adaptive and sustainable disaster management frameworks. The paper concludes with practical and policy-level recommendations for integrating these technologies into Pakistan’s disaster management systems.

Introduction

As climate-related disasters become increasingly common and severe, utilizing cutting-edge innovation for disaster management appears essential [37]. For instance, climate disruption, urbanization, and increased population size in areas susceptible to disasters have all contributed to this sense of need. Consequently, artificial intelligence (AI) is now recognized as a catalyst for innovation in catastrophe preparedness, response, and recovery [46]. AI-enhanced crowdsourcing utilizes machine learning and current social media data to facilitate more effective catastrophe response. This intelligent application of technological innovations not only enables agencies to act quickly but also encourages communities to stay up-to-date, remain connected, and be resilient in the face of disasters. By analyzing public messages, pictures, and videos, AI can recognize crises, assess harm, and direct efforts to save people more quickly than conventional approaches [22].

Artificial Intelligence (AI)

AI enhances disaster management by evaluating real-time information, forecasting dangers, streamlining decision-making processes, and optimizing resource allocation to facilitate more rapid and efficient responses [29]. Thus, among the various AI applications, the most significant involves AI-enhanced crowdsourcing, particularly through social media networks. For instance, after earthquakes, AI bots help coordinate disaster relief operations by extracting survivor locations from social media posts. Additionally, in the context of forest fires, AI models analyze real-time photographs uploaded to social media to monitor the propagation of fires in Pakistan [40]. Moreover, after storms, AI examines comments to recognize vulnerable locations that require immediate assistance. Such digital venues function as immediate data centres, allowing communities, disaster relief agencies, and authorities to work together efficiently during disasters.

Community resilience through social media

Social media has evolved into an essential tool during disasters, helping to enhance community resilience [39]. Social media plays a crucial role in disaster assistance, serving as a key part of individuals and contributing significantly to community resilience. Social media has evolved into a crucial tool during disasters, enhancing community resilience in Pakistan [45]. People can use sites such as Instagram, Facebook, and Twitter to immediately disseminate updates, alerts, and pleas for aid, resulting in an upsurge of real-time, basic-level data. The aforementioned not only maintains individual updates but also promotes communal cohesion and collaborative action.

Integration of AI and social media

The collaboration of machine learning-based analytics and crowdsourced information gathered via social media creates an effective system for detecting critical needs, following crisis development, and aiding immediate decision-making [13]. For instance, AI monitors real-time social media information to recognize crisis patterns, assess damage, and facilitate prompt, informed managerial decisions during a catastrophe. By analyzing enormous quantities of content created by users (which includes messages, postings, photographs, and recordings), AI systems can identify trends, assess damage, and occasionally anticipate potential problems before they occur in Pakistan [42]. It not only enhances contextual understanding but also strengthens local populations by increasing their disaster resistance. When paired with AI-enhanced crowdsourcing, it serves as a powerful emergency preparedness tool, enabling real-time data exchange, faster demand recognition, and synchronised responses. Table 1 illustrates how AI enhances speed, precision, and decision-making over traditional techniques [28].

Table 1.

Conventional versus AI-enhanced method

	Conventional method	AI-enhanced method
Data Collection	Manual Collect data through reports and surveys, manually	Automated Collect data through real-time social media
Decision-making	Reactive Make decisions based on incomplete or insufficient data	Predictive Make decisions based on AI-driven forecasting knowledge
Speed	Slow Its response time is slow because of delayed data	Fast Its response time is fast because of the automated computerized evaluation
Resource Allocation	Generalization Allocate resources through generalized distribution	Optimization Allocate resources through AI-optimized solutions based on specific needs
Accuracy	Human Judgement It depends on human interpretations, which can lead to inconsistencies	Data-Driven It uses data analytics to provide more precise and reliable insights
Communication	Limited It relies on official channels and is often subject to delays	Real-time It enables real-time communication through social media and automated alerts

A game changer for disaster management: AI-enhanced crowdsourcing

As communities evolve into more digital environments, incorporating AI-enhanced crowdsourcing into catastrophe management provides an anticipatory strategy for disaster prevention [55]. For instance, considering an increase in digital interaction, AI-powered crowdsourcing offers immediate information collection, quick responses to emergencies, and community involvement, thereby decreasing the effects of a catastrophe through swift action. AI-enhanced crowdsourcing controls disasters by gathering and evaluating real-time information supplied by the general public via sites such as social media in Pakistan [24]. AI utilizes complex algorithms to rapidly evaluate postings, photographs, and videos, recognizing disasters, estimating damage, and pinpointing areas that require immediate assistance [3]. This enables governments and participants to make much quicker, better-informed choices, increasing the efficiency and accuracy of responding to disasters [43]. There are various AI methods and their applications in responding to catastrophes [47], presented in Table 2.

Table 2.

AI methods and uses

sUses of AI methods in disaster management	AI methods
Offering emergency aid	AI-powered virtual assistance
Evaluating harm from photos or videos	Computer Version
Detecting indications of distress on social media	Natural Language Processing
Forecasting disaster zones	Machine Learning

Pakistan's unique topography and atmospheric conditions expose it to various natural disasters, including storms, accidents, mass movements (such as dry spells), landslides, floods, extreme temperatures, epidemics, earthquakes, and droughts (Hussain et al., 2022). Table 3 and Fig. 1 present the Natural Hazards Occurrence in Pakistan from 1980 to 2020 [53]. These calamities endanger not only individuals and their families but also the nation's infrastructure and emergency response procedures. Recurrent monsoon floods, particularly in shallow and remote regions, have underscored the urgent need for enhanced protection and risk-reduction initiatives [54]. As climate change exacerbates such catastrophes, Pakistan must take more sophisticated, technologically focused ways to protect vulnerable people and build durable resilience.

Table 3.

From 1980 to 2020, average situation of natural disaster occurrence in Pakistan

Category	Incidents
Storm	23
Miscellaneous Accident	37
Mass movement (dry)	2
Landslide	22
Flood	92
Extreme Temperature	15
Epidemic	11
Earthquake	29
Drought	2

Fig. 1.

Natural hazards occurrence in Pakistan, 1980–2020 [53]

Research gap and rationale

As AI and social media are frequently and extensively explored independently in crisis scenarios, a gap remains in recognizing their combined effects on community resilience through crowdsourcing. This gap underscores the need for further research into how AI-powered social media data analysis can facilitate real-time responses and bolster community resilience during disasters [4]. As the quantity and severity of disasters increase, conventional responding methods frequently fail to meet real-time needs [27]. This research is vital as it addresses the urgent need for more intelligent, faster, and community-centred solutions. The present study examines how AI can enhance community resilience through social media-driven crowdsourcing, focusing on the possibilities, challenges, and potential effects on disaster response and preparedness tactics. For instance, highlighting a strong tool to overcome knowledge gaps, stimulate response operations, and encourage communities to act as proactive participants in sustaining themselves during emergencies.

Objectives of the study

This study will explore the following objectives:

1. Identify how AI-enhanced crowdsourcing helps to develop better, stronger communities that survive catastrophic situations.

2. Analyze how AI-powered crowdsourcing could boost the swiftness and precision of response to disasters using social media information.

3. Examine how real-time social media data can help in emergency decision-making processes.

4. Identify the key possibilities and limitations of integrating AI and social media into disaster management strategies.

5. Propose policy implications for incorporating AI-enhanced crowdsourcing techniques into current catastrophe prevention systems.

Following the objectives, to understand the concept of the study, Table 4. Contains the definition of key terms like Community Resilience (CR), Social Media Analytics (SMA), Artificial Intelligence (AI), Crowdsourcing (CDS), AI-Enhanced Crowdsourcing (AIEC), and Disaster Management (DM).

Table 4.

Essential keywords and definitions

Key Word	Definition	Source
Community Resilience (CR)	Community Resilience (CR) refers to a community's ability to anticipate, withstand, and recover from disasters or crises CR is an organization's capacity to predict, survive, adapt to, and recover from catastrophes or emergencies while preserving its fundamental roles and social cohesiveness. A resilient community not only overcomes obstacles but also evolves from them, becoming better organized, more connected, and better equipped to tackle unforeseen hazards. It exemplifies the power of individuals banding together, leveraging regional assets, mutual understanding, and united effort to overcome hardship	[31, 48]
Social Media Analytics (SMA)	Social Media Analytics (SMA) is the procedure of aggregating and analyzing social media data to extract valuable insights SMA is the act of collecting, evaluating, and interpreting data from various social media sites to discover significant findings, current events, and trends. Such knowledge can help disaster managers make faster, better-informed, and more effective responses during emergencies. It examines postings, responses, photos, and collaboration to get insight into public behaviour, viewpoints, and real-time occurrences	[6]
Artificial Intelligence (AI)	Artificial intelligence (AI) is the utilization of technological devices to perform tasks that would normally require human intelligence AI focuses on developing systems that are proficient in performing tasks that would ordinarily require the expertise of individuals. AI enables machines to evolve and respond intelligently to complex situations, making it a valuable tool in various fields, including healthcare, public transportation, crisis management, and others. Such activities involve data analysis, pattern recognition, language comprehension, solving issues, and decision-making	[16]
Crowdsourcing (CDS)	Crowdsourcing refers to the process of obtaining input or services from a large group of people, typically through digital platforms CDS is the process of acquiring thoughts, data, or services from a large number of individuals, typically using internet-based platforms. CDS in disaster management enables the community to exchange current information, report crises, and contribute local insights to response activities. It leverages the community's collective involvement to address issues, generate data, and accomplish tasks more effectively	[25]
AI-Enhanced Crowdsourcing (AIEC)	AI-Enhanced Crowdsourcing (AIEC) is the utilization of AI tools to process, filter, and enhance crowdsourced data for improved real-time decision-making AIEC is the process of combining AI with publicly accessible inputs (like social media postings, photographs, or documents) to increase the effectiveness and precision of data collection and decision-making. By analyzing vast amounts of real-time content that consumers create, AI can identify trends, prioritize essential information, and enable quicker, more intelligent responses in complex scenarios, such as disaster management or emergency response	[22]
Disaster Management (DM)	Disaster management is an organized approach that encompasses preparation, response, rehabilitation, and mitigation activities to mitigate the effects of catastrophes on individuals and infrastructure DM is the organized effort to prepare for, react to, and recover from disasters of any kind. The purpose of disaster management is to assist communities in recovering from disasters as swiftly and securely as possible. It entails planning, setting up resources, mitigating hazards, and ensuring a successful disaster response to safeguard individuals, assets, and the ecosystem	[1]

This study’s objective is to investigate how AI-enhanced crowdsourcing, driven by social media, can improve the efficiency, precision, and collaboration of disaster management efforts, thereby strengthening community resilience.

Materials and methods

This section presents the study design, data source/search strategy, as well as the criteria for exclusion and inclusion. The criteria have now been well-defined to ensure transparency in the discussion on the removal of papers. This involves a focus on language (English), time (2018–2025), geographical orientation (Pakistan/South Asia), and the use of peer-reviewed journal articles, as well as exclusions, including grey literature and non-peer-reviewed sources.

Study design

This work employs a systematic review approach to thoroughly evaluate how AI-enhanced crowdsourcing through social media sites supports disaster management and fosters community resilience. The PRISMA principles were followed to ensure a transparent and organized selection process. This systematic process enables readers to objectively evaluate the reliability and thoroughness of the documentation provided [36]. Employing its criteria and flowchart diagram, authors can explicitly disclose how research has been reviewed and added, thereby reducing risk and enhancing the trustworthiness of the review. It enhances transparency, uniformity, and rigour in the review method. It provides an organized approach for identifying, selecting, and assessing relevant research, thereby enhancing the validity and reliability of outcomes.

Data source/search strategy

To conduct a detailed review of AI-enhanced crowdsourcing that strengthens community resilience through social media, a set of relevant keywords was compiled. A thorough literature review was conducted, leveraging databases such as Google Scholar, Scopus, and Web of Science. Keywords were intentionally selected to narrow down research indexes.

Inclusion and eclusion criteria

Studies were selected on the basis of their relevance, publication quality, and alignment with the study objectives. Moreover, to review prior research published between 2018 and 2025 utilizing a keyword search string presented in Table 5 & Fig. 2. Through thorough screening analysis, the study highlights major trends, real-world implications, and actual gaps in the existing literature. Additionally, numerous research studies across the globe were considered for review. To enhance the clarity and transparency of the systematic analysis process, specific exclusion and inclusion criteria were applied during the selection of literature:

• Language: Only English-published papers were considered to guarantee uniformity and accessibility.

• Time Frame: The review focused on literature published between 2018 and 2025 to capture recent developments and trends in the field.

• Geographic Scope: Preference was given to studies conducted in Pakistan or the broader South Asian region, given the contextual relevance to the research focus.

• Type of Literature: For assurance that the sources were reliable and academically rigorous, only peer-reviewed publications were selected.

• Exclusion Criteria: Publications not written in the English language, grey literature, conference abstracts, and non-peer-reviewed publications were excluded from the analysis.

Table 5.

Search string (Keywords evaluated across distinct international journals, 2018–2025)

Source	String
Google Scholar	“Disaster Management” OR “Risk Management Strategies” OR “Community Resilience” OR “Community Empowerment” OR “Artificial Intelligence” OR “Cutting-edge technologies” OR “AI-Augmented Crowd” OR “Flood Disaster Management Cycle”
Web of Science	“Post-Disaster Resettlement” OR “Disaster Response” OR “Volunteer Crowdsourcing” OR Residents’ Disaster Preparedness” OR “AI-Enhanced Crowdsourcing” OR “AI-driven Risk Management” OR “Multi-Hazard Disasters” OR Natural Hazards”
Scopus	“Disaster Prediction” OR “Rescue Operations” OR “Urban Resilience” OR “Smart Disaster Resilience” OR “Social Media Analytics” OR Analytics of Machine Learning, Crowdsourcing, Disaster Prediction, Reliable Disaster Damage Assessment”

Fig. 2.

Keywords evaluated across distinct international journals (2018–2025)

Ethical considerations and data privacy

Ethical aspects and issues related to data privacy are paramount when it comes to utilizing AI-based and social media-based disaster management solutions. The principles of transparency, consent, and confidentiality must regulate the sociological study of data derived from social media. The crisis response allows all social media posts, so many of its users may not be in a position to know that their information could also be used as part of the crisis response, hence there could be the ability that some information may be used with the aims of abuse by using lack of informed consent to abuse personal integrity.

To reduce these concerns, this paper recognizes the necessity of considering the following issues:

• Anonymization of user data to protect identities,

• Strict compliance with local and international data protection laws, and

• Transparent data handling protocols to ensure responsible AI use.

Without addressing these ethical foundations, the deployment of such technologies may lead to mistrust, public resistance, or even legal repercussions, thereby undermining their intended benefits in disaster response.

Consideration of local socio-cultural contexts

The socio-cultural setting of the place is also instrumental to successful application of AI-augmented crowdsourcing and social media-based tools. In Pakistan, digital literacy is different, and access to mobile internet, as well as the trust punters would grant the technology, have to be considered. For example, there may be situations where some rural communities are more reliant on traditional two-way communication methods and do not actively use social media platforms.

To ensure inclusive and effective use of AI systems, this study emphasizes the importance of:

• Designing culturally relevant communication strategies,

• Conducting community training and awareness programs, and

• Collaborating with local stakeholders who understand regional dynamics and trust networks.

By aligning technology implementation with local socio-cultural realities, these digital solutions are more likely to be acknowledged and chosen by the communities they are intended to serve.

Thematic results/results by keyword grouping

At the start of the search, 2,315 papers were found via 3 separate sources. Following the removal of 1,017 duplicate submissions, the titles and abstracts of the other 1,303 articles were reviewed. 1,075 research papers were removed because they did not fit the study's criteria. The entire contents of the 228 leftover papers have been reviewed for suitability, with 123 papers excluded due to a lack of emphasis on AI-enhanced crowdsourcing or inadequate empirical evidence. Ultimately, 105 papers have been chosen for a thorough evaluation and assessment, as shown in Fig. 3.

Fig. 3.

The comprehensive screening procedure for the most recent AI-enhanced crowdsourcing and disaster management publications

Every year, lots of individuals around the globe are affected by both man-made and natural disasters. Frequently, such disasters result in the loss of human life. Disasters have serious consequences for infrastructure and property, in addition to injuries to people [12].

Stages of disaster management

Disaster management is a systematic method for planning for recovery and responding to climatic or human-made catastrophes [5, 9]. The purpose is to reduce the detrimental effects of catastrophes and create resilient structures that can endure unforeseen circumstances. The four stages of disaster management form an ongoing process that aims to mitigate the detrimental effects of disasters and facilitate a swift recovery, as presented in Fig. 4.

Fig. 4.

Stages of disaster management [38]

The stages of disaster management involve concerted efforts to mitigate hazards, protect individuals and assets, and ensure prompt recovery through effective preparation, efficient asset utilization, and community participation.

The results of the systematic review are presented in the form of the following themes out of the most frequently used keywords: (1) community resilience, (2) social media analytics, (3) artificial intelligence, (4) crowdsourcing, and (5) AI-enhanced crowdsourcing. All these themes are discussed in detail in the following paragraphs.

Community Resilience (CR)

Community resilience is defined as the collective ability of individuals and organizations within a community to organize for, respond to, and recover from disasters or crises [35]. Resilient communities can not only withstand instant shocks, but they are also better prepared for growth and adaptation in the context of upcoming problems [39]. It entails strengthening local communities through the exchange of information, collaborative strategy, and robust networks of assistance that minimize risk. It is the cornerstone of successful disaster planning and recovery, enabling local communities to endure and cope with both short-term and long-term obstacles [33]. Strengthening individuals at the local level, encouraging stakeholder confidence, and developing flexible mechanisms that can adapt to shifting threats and demands are all essential components of building community resilience. The main elements are real-time data, effective management, societal cohesion, effective interaction systems, and equitable decision-making processes [20]. Learning, regional expertise, and access to resources are important in determining how effectively a community recovers after a tragedy. Community resilience refers to a set of environmental, institutional, economic, and social characteristics that collectively enhance a community's ability to cope with disasters [48].

Effect of community resilience on Disaster Management (DM)

Social resources remain the primary process by which community resilience mitigates the effects of disasters and facilitates healing [35]. Advances in resilience evaluation continue to develop the concept and its associated techniques, employing both original and secondary data. Programs around the world are making progress in strengthening resilience through community-based efforts that enhance adaptive abilities. With a rising incidence of all types of weather-related disasters, bolstering communities' resilience and disaster prevention capacity, as well as boosting residents' emergency preparedness, has become a successful way to mitigate risks associated with catastrophes and strengthen residents' welfare [33].

Community resilience in the disaster management of Pakistan

The major catastrophic floods of 2010 in Pakistan’s province (Punjab) caused substantial displacement within nations of rural people, leading to measures to reintegrate the homeless inhabitants into model communities. Governments are less effective than non-governmental organizations (NGOs) in strengthening the whole circumstances of the moved communities [19]. The provision of opportunities for livelihood growth in skills dependent on regional needs, instruction in the maintenance and operation of various MV amenities, and the provision of broad learning opportunities, particularly for women, are all elements in increasing the resilience of NGOs' re-established communities [23].

In Pakistan, community resilience is crucial for lowering people's susceptibility to both environmental and human-caused disasters. In catastrophe-prone areas like Pakhtunkhwa, Balochistan, and Sindh, communities with strong communal bonds, regional expertise, and preemptive disaster readiness recover more quickly and successfully. Resilient communities often engage in proactive notification dissemination, regional risk assessments, and integrated response activities to mitigate the effects of hurricanes, typhoons, and other disasters [34]. Furthermore, community-driven efforts backed by regional nonprofit organizations and federal agencies have effectively raised knowledge, reinforced infrastructure, and developed responsive skills. Yet, the level of resilience varies by location due to socioeconomic inequality, limited educational opportunities, and inadequate societal support. Thus, community resilience is crucial for developing a comprehensive and sustainable catastrophe prevention mechanism in Pakistan [21].

Opportunities and limitations in community resilience implementations

Successful community resilience execution frequently depends on essential integrators, such as solid social connections, regional leaders, access to modern communications and information, and community-based organization participation. Such factors promote collaboration, trust, and information exchange, enabling organizations to better plan for, interact with, and recover from disasters [14]. Despite the growing emphasis on resilience, some obstacles persist. These consist of scarce funds, a shortage of knowledge or catastrophic learning, uneven technological access, and inadequate institutional assistance. Furthermore, misrepresentation during emergencies, inadequate stakeholder cooperation, and socioeconomic status inequality can hinder collaborative efforts, making it challenging to sustain resilience projects over time [44]. Table 6 explains the study, its key findings, and implications for community resilience.

Table 6.

Community resilience key findings

Study	Main findings	Implications
[35]	Community resilience minimizes the impact of disasters	Programs worldwide are making headway in improving resilience by leveraging community-based activities to enhance adaptive qualities
[33]	The citizens' disaster readiness worsens as the percentage of community resources increases	Establishing resilient disaster mitigation strategies in disaster-prone communities, as well as developing policies that strengthen residents' abilities to safeguard themselves from catastrophes
[23]	Increasing community resilience during post-catastrophic resettlement reduces risk	Authorities and community organizers should address current weaknesses by engaging the business community and considering social and economic factors when establishing communities
[34]	In Pakhtunkhwa, Balochistan, and Sindh, community resilience helps reduce society’s vulnerability to both man-made and natural calamities	Regularly evaluate the usefulness of crowdsourcing techniques. Feedback methods have the potential to alleviate difficulties in disaster management
[21]	Community resilience is of vital importance for developing a comprehensive and sustainable disaster prevention strategy in Pakistan	Community-driven efforts, supported by regional nonprofit organizations and federal agencies, have effectively increased knowledge, strengthened infrastructure, and developed responsive skills
[14]	Successful community resilience execution frequently recovers from catastrophes	Encourage collaboration, faith, and information exchange, enabling organizations to better plan for, respond to, and recover from emergencies
[44]	False statements during emergencies, insufficient stakeholder engagement, and socioeconomic status inequality can hinder collaborative efforts, making it difficult to sustain resilience programs over time	Coordinate with innovation businesses, educational settings, and NGOs to establish and sustain strong crowdsourcing mechanisms

Social Media Analytics (SMA)

Social media analytics is a systematic method of monitoring, collecting, and evaluating information generated by social media sites to gain important insights [6]. In disaster management, social media analytics can generate pre-emptive alert warnings, pinpoint damaged locations, and aid in effective decision-making by converting unstructured internet posts into actionable intelligence. This method enables companies to track discussions in real-time, identify prominent themes, and measure public reaction to specific events or emergencies [39]. It entails analyzing multiple types of content, including text, hashtags, shares, videos, photographs, and comments left by users, to comprehend trends, public mood, and newly occurring problems.

Effect of social media analytics on disaster management

Despite the wide range of information fields in social media data, the following categories (network, content, time, and space) have received special focus for mining usable data to increase situational consciousness and disaster preparedness. Thus, social media analytics has gained prominence in natural disaster management [50]. The primary concern of the study field is the use of data from social networks for disaster response to unforeseen events. However, social media-based information catastrophe response to catastrophes caused by humans is also a growing area of research attention. Social media analytics in disaster response primarily focus on community intelligence, geographical awareness, and situational awareness [6].

Social media analytics in the disaster management of Pakistan

Social media analytics have become a recognized effective technique for enhancing disaster preparedness in Pakistan. During an emergency, mediums like Facebook and Twitter function as swift data exchange paths, allowing administrators to track current public opinion, detect critical needs, and monitor the regional extent of catastrophes. By analyzing patterns, search terms, and location-tagged content, decision-makers can receive immediate information on regions impacted, assess damage levels, and more effectively concentrate remediation actions [10]. In Pakistan, where official buildings are frequently affected during disasters such as hurricanes and quakes, SMA fills crucial knowledge gaps. It also allows citizens to submit personal narratives, promoting reciprocal interaction among ordinary people and crisis management institutions. Although the efficacy of such technologies depends on technological proficiency, the availability of technology, and the capacity for filtering disinformation. Despite these obstacles, SMA has enormous potential for enhancing emergency preparedness and enabling flexible, data-driven recovery strategies nationwide [45].

Opportunities and limitations in social media analytics implementations

Social media offers significant opportunities for disaster management improvement by facilitating swift data exchange, enabling immediate interaction, and promoting civic participation. It enables governments to issue immediate notifications, assess public sentiment, and collect on-the-ground information from those affected. Furthermore, social media platforms can foster cooperation among government organizations, non-governmental organizations (NGOs), and the general public, leading to a broader and more flexible disaster management framework [15]. The incorporation of AI expands these prospects by enabling faster data processing and informed decision-making during disasters. While social media analytics can provide significant current information throughout disasters, there are numerous limitations to its deployment. Data precision and dependability remain key challenges, as user-submitted content may contain disinformation or be incomplete. Furthermore, unequal availability to the web and technological proficiency may eliminate disadvantaged groups, resulting in biased statistics. Privacy constraints, ethical considerations, and the lack of uniform statistical models all hamper the efficient implementation of social media analytics in disaster management. These problems underline the importance of integrating these instruments into rescue plans with caution and responsibility [32]. Table 7 explicates the study, its key findings, and implications for social media analytics.

Table 7.

Social media analytics key findings

Study	Main findings	Implications
50	Social media analytics upgrades natural disaster management	Researching potential and obstacles in combining information from social networks with official records, such as census and sensor data from remote areas, is beneficial for natural disaster management
[6]	Social media analytics is a burgeoning field that improves disaster management	Decision-making procedures, emergency response standards, architectural instructions, and urbanization strategies can all be evaluated through social media analytics based on past catastrophes
[10]	Social media analytics through mediums like Facebook and Twitter provide a more flexible disaster management framework	By assessing patterns, search terms, and location-tagged content, decision-makers can receive immediate information on regions impacted, assess damage levels, and better concentrate remediation actions
[45]	SMA fills crucial knowledge gaps and has enormous potential for boosting emergency preparedness in Pakistan	Cooperation among government organizations and non-governmental organizations (NGOs) improves the disaster management structure
[15]	Allows governments to deliver notices instantly, measure public sentiment, and gather on-the-ground information from citizens affected	Educational initiatives can help communities offer precise and up-to-date data during catastrophes
[32]	SMA provides timely and accurate information throughout disasters	Privacy concerns, ethical considerations, and a lack of uniform statistical models allow governments to issue alerts quickly, analyze public sentiment, and collect on-the-ground information from those affected

Artificial Intelligence (AI)

Artificial intelligence (AI) is a field of technological advancement that emphasizes developing systems and devices accomplished of replicating human cognitive abilities [16]. AI is transforming the way technology works globally by automating mundane tasks and facilitating sophisticated decision-making in critical fields, such as healthcare, transportation, and emergency response. AI employs methods such as machine vision, natural language processing, and algorithmic learning, enabling machines to analyze data, handle new resources, and improve their effectiveness over time. These systems are designed to perform functions such as acquiring expertise, detecting patterns, understanding language, making judgments, and resolving issues without being customized for each scenario [51].

Effect of AI on Disaster Management (DM)

Artificial intelligence techniques are useful in improving disaster management. It is determined that present patterns are focused on catastrophe response and mitigation. The field of geography tends to advance, providing a timely answer and diverse goals for Gastrointestinal Science. Thus, it is necessary to establish and consider the geographical implications. Currently, geographical information systems and remote sensing are powerful technologies that provide a novel approach to understanding situations [2]. AI has enhanced catastrophe management by improving prediction, enhancing response, and mitigating economic and ecological harm. Technologies based on artificial intelligence enable cities to transition from responsive catastrophe response to proactive risk reduction by leveraging real-time information interpretation, sophisticated resource allocation, and automation [11].

AI in the disaster management of Pakistan

Artificial intelligence (AI) is enhancing emergency management in Pakistan by improving the effectiveness and accuracy of disaster preparation, recovery, and response plans. AI-enabled mechanisms may evaluate massive volumes of information, including climate data, aerial photographs, and online conversations, to forecast and track natural disasters such as hurricanes, earthquakes, and collapses. In areas where alert systems are inadequate, AI models can send timely alarms and direct evacuation strategies, minimizing the loss of assets and lives [42]. In addition, AI-powered technologies aid in evaluating harm and allocating resources by rapidly analyzing current information, enabling quicker and more informed decisions. In Pakistan, where catastrophes frequently strike distant and underserved populations, AI can help address significant communication and collaboration barriers. Although AI's greatest potential has yet to be fully realized due to obstacles, including inadequate technological support, data accessibility, and the need for skilled individuals. Despite these challenges, utilizing AI in emergency preparedness systems is a promising phase towards developing a more robust and responsive catastrophe management framework in Pakistan [49].

Opportunities and limitations in AI implementations

Artificial intelligence presents disruptive opportunities in disaster management by enabling faster, data-driven decision-making and enhancing systems for early detection. AI can recognize forecasts and potential threats by analyzing massive amounts of current information from diverse sources, like social media, satellite imagery, and sensors. It also promotes effective allocation of resources, enhances interaction among stakeholders, and tailors reaction plans to meet community requirements. When properly integrated, AI can significantly enhance catastrophe preparedness, intervention, and rehabilitation initiatives, creating more adaptable and robust networks [16, 52]. Although it holds tremendous promise, the utilization of artificial intelligence (AI) in disaster management has several limitations. One significant problem is the requirement for comprehensive, high-quality information sets, which are often inaccessible or unreliable in regions prone to disasters. AI networks may potentially suffer from situation determination, resulting in erroneous projections or conclusions. Ethical considerations, including confidentiality of information and algorithmic prejudice, impede its implementation. Furthermore, inadequate technological facilities and a scarcity of experienced personnel in certain locations hinder the successful integration of AI into disaster recovery plans [51]. Table 8 explains the study, its key findings, and implications for artificial intelligence.

Table 8.

Artificial intelligence key findings

Study	Main findings	Implications
[2]	Artificial intelligence technologies can help improve disaster management	Graphic representation skills, satellite imagery, and AI assessment can help authorities make swift judgments following catastrophic events
[11]	AI can enhance disaster management by mitigating economic and human losses and promoting sustainable urban development	By leveraging AI and global collaboration, cities can create communities that are better equipped to face foreseeable weather and disaster problems
[42]	Artificial intelligence (AI) is transforming the response to emergencies in Pakistan	Support collaboration among authorities, business communities, and educational sectors to promote creativity and exchange resources when creating AI-powered emergency preparedness systems
[49]	AI can help address key barriers to interaction and collaboration in disaster management	Provide disaster mitigation workers with specific instructions to help them become experts in AI tools and technologies that address disaster-related problems
[16, 52]	AI may dramatically improve catastrophe preparedness. AI-powered technologies help with evaluating harm and allocating resources, creating networks more adaptable and robust	Creating robust data gathering and leadership mechanisms to enable AI models is essential for the efficacy of AI-enhanced catastrophe recovery systems
[51]	The successful incorporation of AI into catastrophic recovery plans can result in erroneous projections of catastrophe recovery systems	Integrating AI practices into current disaster management policies can enhance alert systems, threat assessment, and asset allocation during disasters

Crowdsourcing (CDS)

Crowdsourcing is an interactive method of collecting thoughts, data, or services from an extensive number of individuals, typically via online mediums. Crowdsourcing in crisis circumstances enables people to provide on-the-ground updates, report risks, and aid agencies in real-time, making it a crucial tool for enhancing contextual awareness and responding to operations. It draws on the insights, experiences, and input provided by others in the community to complete activities that would be too difficult or time-consuming for one entity [25].

Effect of CDS on Disaster Management (DM)

Crowdsourcing and Community engagement are effective ways to boost disaster mitigation and rescue efforts following the huge earthquake that struck Syria, and strengthen responses to catastrophes and their results. Over the past decade, smartphone applications, SMS-based frameworks, and social media sites have been utilized to gather information regarding regions impacted, damage to infrastructure, and emergency requirements, helping governments and NGOs better prioritize response operations [8].

CDS in the disaster management of Pakistan

Crowdsourcing is emerging as an effective way to enhance disaster prevention measures in Pakistan, particularly in areas prone to natural disasters such as storms, earthquakes, and floods. Crowdsourcing fills important gaps in conventional response methods, which are often delayed or limited by resources. Crowdsourcing promotes involvement among communities, rendering DM more equitable and participative. In Pakistan, where distant communities typically receive limited access to formal media, crowdsourced information may play a crucial role in empowering those who are often overlooked. Although the correctness of the information and the technological gap are still major challenges. Despite these obstacles, crowdsourcing has enormous potential to supplement conventional preparedness methods and enhance resilience [30].

Opportunities and limitations in CDS implementations

Crowdsourcing offers tremendous prospects for disaster management by enabling instantaneous information to be gathered directly from harmed populations. It enables citizens to provide initial insight, so increasing awareness of circumstances and facilitating speedier, personalized decision-making. Such an interactive strategy encourages involvement from communities, builds confidence among individuals and authority, and enables a more focused allocation of resources. Furthermore, when integrated with online mediums, crowdsourcing can help overcome knowledge disparities, particularly in distant or underdeveloped regions, enabling emergency responses to be more affordable and adaptable [25]. While crowdsourcing increases community involvement in handling emergencies, its implementation has significant limits. The dependability and quality of user-generated information can be questioned, particularly in chaotic settings. A scarcity of ways to verify information may contribute to the spread of disinformation. In addition, discrepancies in internet availability and literacy might lead to the removal of disadvantaged communities, resulting in data deficits. Coordinating and handling massive amounts of crowdsourced material requires a robust technology infrastructure and skilled individuals, which may not always be readily available [26]. Table 9 explains the study, its key findings, and implications for Crowdsourcing.

Table 9.

Crowdsourcing key findings

Study	Main findings	Implications
[8]	Crowdsourcing is an effective way to aid in disaster and recovery efforts	Crowdsourcing enables communities to offer real-time data, resulting in more rapid and specific replies
[30]	In Pakistan, crowdsourcing promotes involvement among communities, rendering DM more equitable and participative	Develop mechanisms for implementing authenticated crowdsourcing data into disaster prevention authorities' business processes
[25]	Crowdsourcing can help overcome knowledge disparities, particularly in distant or underdeveloped regions, thereby enabling more effective emergency responses	Enhancing community understanding and proficiency in technology can help communities provide precise data during disasters
[26]	Crowdsourcing increases community involvement in responding to emergencies	Develop open and equitable crowdsourcing platforms for more effective decision-making during disasters

AI-Enhanced Crowdsourcing (AIEC)

AI-enhanced crowdsourcing integrates the general population's collaborative information with the powerful reasoning abilities of AI to tackle complicated problems more effectively. AI enhances the speed, precision, and flexibility of information assessment, making it particularly useful in time-sensitive situations, such as disaster response, where prompt findings can save lives and assets [22]. It employs sophisticated algorithms to discover patterns, identify abnormalities, and enable real-time decision-making. In contrast to conventional crowdsourcing. This technique leverages massive amounts of content generated by individuals, including reports, photographs, or social media updates, and analyses it [41].

Opportunities and limitations in AMEC implementations

AI-enhanced crowdsourcing has enormous potential for enhancing catastrophe management through the integration of individual expertise with automated efficiency [18]. It enables the rapid examination of massive amounts of consumer-generated data, leading to enhanced situational awareness and greater precision in decision-making. Such coordination increases the speed and accuracy of disaster action, aids in identifying critical requirements, and facilitates resource allocation based on actual basic levels data. Furthermore, it promotes community involvement while decreasing the traditional strain on emergency preparedness groups, resulting in stronger and more flexible emergency preparedness structures [17]. Despite AI-enhanced crowdsourcing assisting crisis management by streamlining data evaluation and decision-making, it has several drawbacks. Inaccurate or biased user-created material might confuse AI algorithms, particularly in the absence of robust verification processes. The integrity and accessibility of data vary greatly, influencing the effectiveness and trustworthiness of AI systems. Mechanical difficulties, such as insufficient facilities, high implementation costs, and the need for experienced staff, may hinder implementation, especially in low-income countries [7]. Furthermore, concerns about the accuracy of information, legal usage, and algorithmic transparency create substantial impediments to confidence and widespread acceptance.

Discussion

The objective of this research endeavor is to examine how AI-enhanced crowdsourcing, facilitated by social media, can improve the effectiveness, precision, and collaboration of disaster management activities, thereby strengthening community resilience. In Pakistan, as in other countries in the Global South, the growing reliance on digital resources reflects a global trend of leveraging technology to accelerate and refine disaster responses. AI-enhanced crowdsourcing has proven especially effective in harnessing real-time social media information, enabling more accurate and timely decision-making during disasters. By situating Pakistan’s experience alongside international efforts, this study underscores both shared opportunities and context-specific challenges in integrating AI into disaster management practices. Community resilience is shown to improve when digital platforms are used not only for information dissemination but also for active community engagement and feedback during disaster events.

However, several contradictions and research gaps were identified. While artificial intelligence and social media analytics are well-researched individually, their combined effect on disaster management, particularly through crowdsourcing mechanisms, remains underexplored. This confirms the central research gap: the lack of studies examining the integration of AI and social media to enhance community resilience through crowdsourcing in disaster-prone contexts, such as Pakistan.

Limitations of the review

Some studies highlight the benefits of AI in automating emergency responses, yet others caution against over-reliance on unverified social media data, pointing to potential accuracy and trust issues. There is also a lack of uniform frameworks for integrating these technologies into formal disaster management systems, reflecting both a gap in practice and policy ambiguity. When addressing the research questions, it is evident that the considerations based on the research findings suggest that AI-enhanced crowdsourcing can indeed facilitate the creation of more adaptive and responsive communities. It has the ability to do that by supplying localized and real-time information, thereby enhancing awareness of a situation. AI also improves the accuracy and the rate at which this information can be interpreted making decisions faster and informed. Moreover, the contemporary use of social media provides relevant clues about the developing crisis, which guide authorities in allocating resources and setting priorities. Despite its potential, there are limitations to consider, including the problem of data reliability, algorithm bias, inadequate infrastructure, and data privacy concerns.

Practical applications

Real-time crisis monitoring

The use of AI can analyze high volumes of social media data to locate emergencies, map affected areas, and identify needs in real-time.

Community engagement platforms

Crowdsourcing platforms allow citizens to share ground-level information, report incidents, and access updates, enhancing two-way communication during disasters.

Automated early warning systems

AI-driven models can analyze trends and trigger automated alerts through social media and mobile apps, enabling faster community response.

Damage assessment and recovery planning

Following a disaster, AI can aid in assessing damage by utilizing crowdsourced images and geotagged reports to inform effective recovery and rebuilding strategies.

Target resource allocation

Predictive analytics can help authorities allocate emergency resources more efficiently by identifying high-risk zones and prioritizing response efforts.

Support for decision-makers

By converting unorganised social media information into useful insights, AI enhances situational awareness for disaster management agencies and policymakers.

Recommendations for future research or policy

The limitations mentioned above must be addressed to fully integrate AI and social media into disaster management procedures in Pakistan. Implications of the findings suggest the need for:

• Policy frameworks that sustain the ethical and secure use of AI and social media in emergencies.

• Capacity building among disaster response teams to interpret and act on digital data.

• Community training programs to encourage responsible participation in digital platforms during crises.

Incorporating AI-powered crowdsourcing technologies into emergency response strategies can increase rapid decision-making and contextual awareness. Authorities ought to encourage this incorporation. Appropriate standards are necessary to promote the responsible use of social media during disasters, ensuring that real-time community feedback can be efficiently processed and validated by AI methods. To improve community engagement in electronic disaster mechanisms, policymakers and organizations should implement comprehensive policies that encourage communities to participate effectively in crowdsourcing networks. This would boost regional resilience and prior alerting skills.

Policies must address the confidentiality of data, authorization, and bias in AI disaster response mechanisms, to ensure that accessible data is used fairly and transparently. Spend on training, proficiency with technology, and infrastructure to support the efficient utilization of AI-enhanced disaster management solutions by participants and residents. Future research can explore the development of more adaptable artificial intelligence models that can process real-time social media information while also understanding the psychological mood and immediacy of content created by individuals during disasters. Moreover, to ensure the secure and accountable use of AI in disaster management, the study may focus on addressing ethical issues, including the confidentiality of information, disinformation, and equal access to information. Furthermore, combining AI-enhanced crowdsourcing with regional disaster relief services and smartphone applications could result in improved accessibility and adaptability structures, particularly in underserved or isolated areas. Future work should aim to develop all-in-one models that integrate SMA and AI in disaster management, with a focus on scalability, inclusivity, and contextual relevance.

Conclusion

This paper aims to discuss the potential role that artificial intelligence-enhanced crowdsourcing, aided by social media analytics and AI tools, can play in enhancing disaster management in Pakistan by increasing community resilience. This study offers a comprehensive evaluation of how AI-enhanced crowdsourcing through social media can enhance disaster management by increasing community resilience, accelerating disaster preparedness, and identifying significant problems and potential issues related to online disaster cooperation in Pakistan. The research found that combining AI with real-time social media data reduces response time, improves the accuracy rate, and enhances overall response capabilities to a crisis. Combined with social media, AI-enhanced crowdsourcing offers a robust approach to disaster management. It enhances the adaptability and resilience of injured communities by enabling the real-time collection of information, rapid assessment, and community engagement. This research highlights a critical gap, the limited understanding of how the combined use of AI and social media influences community resilience. It underscores the transformative potential of digital tools to revolutionize disaster management by integrating technology with citizen participation, thereby closing knowledge gaps and facilitating timely, evidence-based decision-making. As disasters become increasingly complex, adopting intelligent, community-centered, and adaptive responses is no longer optional but essential for building safer and more resilient societies. In the case of Pakistan, these findings resonate with broader experiences across the Global South, where similar efforts to harness AI-driven crowdsourcing are reshaping disaster preparedness and response. By situating Pakistan within this comparative context, the study provides valuable insights for policymakers, emergency responders, and researchers seeking to modernize disaster management systems in vulnerable regions worldwide.

Authors’ contributions

Conceptualization, S.K.A. and U.N.; methodology, S.K.A., R.R and N.S.B.K; writing, S.K.A., R.R., U.M., and N.S.B.K.; supervision, R.R., N.S.B.K. and S.K.A.; project administration, S.K.A., and U.N. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data availability

The original contributions presented in the study are included in the article and supplementary material. Further inquiries can be directed to the corresponding author.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.