Transformation of agricultural extension services in the digital era: Opportunities and challenges

smallholder, women and remote farmers who are least able to access alternative information sources. The past two decades have witnessed an unprecedented expansion of information and communication technology (ICT) infrastructure in rural areas of the developing world, driven principally by the diffusion of mobile telephony and, more recently, mobile internet and smartphone penetration (Khan et al., 2025; Ndimbo et al., 2025) [27,36] . This has catalyzed the emergence of a distinct sub-field of digital or eextension, encompassing mobile-based advisory platforms, interactive voice response systems, call centres, SMS-based alerts, web and mobile applications, remote sensing and geospatial decision-support tools, the Internet of Things (IoT), blockchain-enabled traceability and, most recently, generative artificial intelligence (AI) chatbots capable of delivering conversational, personalized advisory services at scale (Mulungu et al., 2025; Neeraja et al., 2026; Roy et al., 2026) [31,37,46] . The COVID-19 pandemic served as a further inflection point, compelling extension systems worldwide to pivot rapidly toward remote and hybrid delivery modalities and accelerating institutional acceptance of digital tools that had previously been treated as peripheral innovations (Singh et al., 2025; Hassoun et al., 2023) [53,23] . Digital transformation in agricultural extension is, however, neither uniform nor unambiguously beneficial. While digital tools promise to overcome long-standing constraints of scale, cost and timeliness, they simultaneously introduce new risks relating to the digital divide, data governance, content quality, and the potential marginalisation of farmers who lack access to devices, connectivity or digital literacy (Ragetlie et al., 2022; Ahad et al., 2024) [41,3] . A comprehensive assessment of the opportunities and challenges associated with this transformation is therefore both timely and necessary. This review accordingly (i) traces the historical evolution of agricultural extension services; (ii) analyses the structural limitations of traditional extension approaches that have motivated the search for digital alternatives; (iii) catalogues the principal digital technologies now being deployed in extension and advisory delivery; (iv) evaluates the opportunities and constraints associated with digital extension; (v) examines institutional readiness and policy frameworks required to support digitalisation; (vi) presents comparative case evidence from selected countries; and (vii) identifies priority directions for future research.

Evolution of Agricultural Extension Services

The term 'extension' originated in nineteenth-century England to describe university education extended beyond the campus to the wider public, and the concept was subsequently adapted to agriculture in the United States through the Smith-Lever Act of 1914, which formally established the Cooperative Extension Service within the land-grant university system (Al-Kaisi et al., 2015) [5] . This model, built on a linear transfer-of-technology paradigm in which research generated at experiment stations was disseminated to farmers through trained agents, was subsequently exported to much of the colonized and newly independent world during the twentieth century (Rogers, 1988) [44] . In the 1970s and 1980s, the World Bank promoted the Training and Visit (T&V) system as a standardized extension model across dozens of countries, characterized by a fixed calendar of agent visits to contact farmers, a single line of technical command, and regular inservice training of extension staff (Feder et al., 1987) [15] . Empirical evaluations of T&V produced mixed results: studies in northwest India reported measurable wheat yield gains attributable to T&V extension (Feder et al., 1987) [15] , while assessments in rainfed regions of Ethiopia and elsewhere found that the system's reliance on a narrow cadre of contact farmers limited its reach to the wider farming population and that women were largely excluded from its structures (Rokvić & Vaško, 2016) [45] . By the late 1990s, the recurrent fiscal burden of T&V, its top-down orientation, and its poor fit with heterogeneous farming systems had prompted a broad institutional retreat from the model. The following two decades saw a diversification of extension approaches, including participatory methods such as farmer field schools, decentralized and pluralistic advisory systems involving non-governmental organisations and private input suppliers, farmer-to-farmer and demanddriven extension, and increasing emphasis on market linkage and value-chain facilitation rather than production advice alone (Khan et al., 2025) [27] . This period also saw growing interest in the role of ICT: early applications, including radio broadcasts, video vans and rural telecentres, provided the first non-traditional channels for reaching dispersed farming populations, laying the groundwork for the mobile-and internet-based systems that followed (Khan et al., 2025) [27] . A systematic review of the evolution of ICT in agricultural extension over the past seven decades identifies a clear shift from one-way, linear dissemination models toward two-way, farmer-centred and co-innovationoriented communication, driven successively by radio, television, the internet, mobile telephony and, most recently, smartphone-enabled social media and artificial intelligence (Khan et al., 2025) [27] . This trajectory reflects a broader reconceptualization of extension itself-from a mechanism for transferring predetermined technical packages to a facilitative, networked function that connects farmers, researchers, input suppliers, financial institutions and markets within an agricultural innovation system (Davis et al., 2025). The digital era represents the latest, and arguably most disruptive, phase in this long institutional evolution, one in which the extension agent is increasingly complemented, and in some functions substituted, by digital intermediaries. Table 1 summarizes this trajectory as a timeline of successive extension paradigms, their defining characteristics, and their principal limitations. [5,44] 1970s [15,45] 1990s-2000s

Participatory and pluralistic extension Farmer field schools, NGO and privatesector involvement, demand-driven advice

Fragmented coordination; uneven institutional funding (Khan et al., 2025) [27] 2000s-2010s

Traditional Agricultural Extension: Limitations and Challenges

Notwithstanding its historical contributions, the conventional face-to-face extension model exhibits several structural limitations that have become increasingly apparent as farming populations have grown and diversified. First, extension-worker-to-farmer ratios in many developing countries remain far below recommended norms, so that even well-designed programmes can reach only a small fraction of farm households within a season (Steinke et al., 2021) [54] . Second, the T&V and similar top-down models were frequently criticized for weak downward accountability, limited farmer participation in programme design, and messages that were poorly adapted to local agroecological and socio-economic conditions (Feder et al., 1987) [15] . Third, traditional extension services have historically struggled to maintain robust linkages with agricultural research systems, resulting in a persistent gap between technologies generated at research stations and those actually recommended to farmers, compounded by inadequate numbers of subject-matter specialists able to translate research findings into field-ready advice (Feder et al., 1987) [15] . Fourth, recurrent cost structures dependent on donor or government financing have proven fiscally unsustainable in many contexts, leading to abrupt curtailment of services once external funding lapsed (Rokvić & Vaško, 2016) [45] . Fifth, face-to-face extension is intrinsically constrained in its ability to deliver timesensitive information-such as pest outbreak alerts, weather warnings or market price movements-at the speed required for farmers to act effectively, a limitation that becomes acute during crises such as pandemics, floods or locust invasions that restrict physical movement and contact-based service delivery (Mulungu et al., 2025; Singh et al., 2025) [31, 53] . Sixth, and of particular significance for equity, women farmers have historically been under-served by traditional extension structures, which have tended to interact preferentially with male household heads and 'progressive' or contact farmers, thereby reproducing and, in some contexts, widening gender gaps in access to agricultural knowledge and inputs (Rokvić & Vaško, 2016; Ahad et al., 2024) [45,3] . Finally, traditional extension systems have often relied on generalized, one-size-fits-all technical packages disseminated through top-down communication, an approach poorly matched to the high heterogeneity of smallholder farming systems and to farmers' own profit and risk considerations, which do not always align neatly with yield-maximising recommendations derived from experimental research plots (Steinke et al., 2021) [54] . These cumulative limitations of reach, cost, timeliness, accountability, gender inclusion and message relevance constitute the principal rationale for the search for complementary digital delivery mechanisms capable of overcoming the scale and cost constraints inherent in purely human-mediated extension.

Digital Technologies Transforming Agricultural Extension

A broad and rapidly evolving suite of digital technologies is now being deployed across the extension value chain, ranging from simple voice-and text-based mobile services to sophisticated artificial-intelligence-driven decisionsupport systems. These technologies can be broadly grouped into six categories: mobile-based advisory services; eportals and mobile applications; remote sensing, geospatial and precision-agriculture tools; the Internet of Things and sensor networks; blockchain-enabled data and supply-chain systems; and artificial intelligence, machine learning and generative chatbot-based advisory systems.

Mobile-Based Advisory Services

Mobile telephony remains the most widely adopted digital channel for agricultural advisory delivery in the developing world because of its low cost, high penetration and compatibility with basic-feature handsets (Gouroubera et al., 2025; Khan et al., 2020) [20,26] . Services in this category include voice-based interactive voice response (IVR) systems, farmer hotlines and call centers, SMS-based weather and price alerts, and voice or text advisories tailored to specific crops. India's Kisan Call Centre, established in 2004, exemplifies this model, providing farmers with tollfree, local-language access to agricultural experts, while similar national-level 'farmer hotlines' have since been established in Ethiopia and elsewhere (Oumer, 2025; Singh, 2015) [38,52] . Meta-analytic evidence indicates that mobilephone-mediated advisory services can meaningfully improve technology adoption, though effects on yield and profitability vary considerably by context and delivery design (Gouroubera et al., 2025; Giulivi et al., 2023) [20,19] .

E-Portals and Mobile Applications

Web-based e-portals and smartphone applications now provide farmers with crop-specific advisories, market price information, soil health data, government scheme information and input-marketing linkages within a single interface (Mane et al., 2023) [29] . In India, platforms such as the Farmers' Portal and privately developed mobile applications integrate weather forecasting, crop advisory and market intelligence functions, while systematic reviews across South and Southeast Asia identify usability, perceived usefulness and trust as key determinants of sustained farmer engagement with such applications (Adesiji et al., 2024; Sen et al., 2024) [2,49] .

Remote Sensing, Geospatial and Precision-Agriculture Tools

Satellite and drone-based remote sensing, combined with geographic information systems, enable field-level monitoring of crop health, soil moisture and irrigation requirements, allowing extension providers to generate sitespecific rather than generic recommendations (Gupta & Pal, 2025; Sai Satish et al., 2025) [22,48] . Precision agriculture decision-support tools that integrate weather, soil and cropgrowth models have been used to optimise irrigation scheduling and input application, reducing costs while improving resource-use efficiency (Narendar et al.,2026; Delgoda et al., 2016) [34,13] .

The Internet of Things and Sensor Networks

IoT-enabled sensor networks provide continuous, real-time data on soil moisture, temperature, humidity and livestock health, feeding automated advisory and decision-support systems and reducing the reliance on periodic manual field assessment (Friha et al., 2021; Subeesh & Mehta, 2021) [17,55] . Applications extend to livestock monitoring through wearable biosensors (Alipio & Villena, 2023) [4] and to automated irrigation control informed by weather forecasts and crop-water models (Delgoda et al., 2016) [13] , with IoT increasingly forming the sensing layer that supplies data to higher-level AI and blockchain systems (Friha et al., 2021) [17] .

Blockchain-Enabled Data and Supply-Chain Systems

Blockchain technology is being explored primarily for its capacity to ensure transparent, tamper-resistant recordkeeping across agricultural supply chains, supporting traceability, quality assurance and trust between smallholder producers and downstream buyers (Pal & Kant, 2019; Rehman et al., 2023) [39,43] . Empirical feasibility assessments in African food supply chains and broader technical surveys of blockchain-IoT applications in agriculture and livestock management corroborate this traceability potential while highlighting significant implementation barriers (Tang et al., 2024; Yang et al., 2025) [56,58] . Combined with IoT sensing, blockchain has been proposed as a mechanism for secure agricultural data management and for underpinning climate-smart agriculture information systems, although its application within extension and advisory services specifically remains at an early, largely pilot stage (Safeer et al., 2024; Dey & Shekhawat, 2021) [47,14] .

Artificial Intelligence and Generative Chatbot-Based Advisory

The most recent and rapidly advancing frontier of digital extension involves artificial intelligence, machine learning and, since 2023, large language model (LLM)-based generative AI chatbots capable of providing personalized, conversational advisory responses in local languages (Neeraja et [37,46,24] . Systems such as Digital Green's Farmer. Chat combine curated, government-vetted agronomic knowledge bases with LLM-driven natural-language interfaces, allowing farmers to pose open-ended questions and receive contextually tailored responses regarding crop selection, pest management and input use (High et al., 2026; Feyisa, 2020) [24,16] . Machine-learning-based crop disease diagnosis using image recognition, AI-driven yield prediction and NLP-based advisory chatbots are increasingly piloted in India and sub-Saharan Africa, although evidence on their reliability, farmer trust and impact at scale remains preliminary (Neeraja et [37,25,12] . A summary of the principal digital technology categories, representative tools and their extension applications is presented in Table 2. [37,46,24]

Opportunities Presented by Digital Extension

Digital technologies present several substantive opportunities for revitalizing agricultural extension. The most fundamental is scale: mobile-and internet-based advisory services can, in principle, reach millions of farmers simultaneously at a fraction of the marginal cost of face-to-face contact, overcoming the extension-worker-to-farmer ratio constraint that has historically limited coverage (Gouroubera et al., 2025; Ndimbo et al., 2025) [20,36] . Because digital channels are not bound to a fixed visitation schedule, they also enable near real-time delivery of timesensitive information such as pest and disease alerts, weather warnings and market price movements, allowing farmers to act promptly on emerging risks and opportunities (Yegbemey & Egah, 2021; Gouroubera et al., 2023) [59,21] . Second, digital tools enable a degree of personalization that generic, campaign-style extension messaging cannot easily achieve. AI-driven advisory systems can tailor recommendations to a farmer's specific location, crop, farm size and even growth stage, moving advisory delivery away from blanket recommendations toward genuinely sitespecific guidance (High et al., 2026; Yegbemey & Egah, 2021) [24,59] . Third, digital platforms facilitate two-way and farmer-to-farmer communication, supporting peer learning, feedback loops and co-innovation in ways that traditional one-way broadcast and even T&V-style visits could not readily support (Khan et al., 2025; Mulungu et al., 2025) [27,31] . Fourth, digital extension can improve resilience and continuity of service delivery during crises that restrict physical movement, as demonstrated during the COVID-19 pandemic, when extension systems worldwide pivoted to mobile applications, radio and virtual engagement to sustain farmer contact (Singh et al., 2025; Hassoun et al., 2023) [53,23] . Fifth, when deliberately designed with genderresponsive features-including content in local languages, household-level rather than individual targeting, and involvement of women in application design-digital extension has shown potential to narrow, rather than widen, gender gaps in access to agricultural information (Ragetlie et al., 2022; Mutuku et al., 2026) [41,33] . Sixth, integration of digital advisory with market information and mobile financial services can strengthen farmers' bargaining position and reduce dependence on intermediaries, as illustrated by platforms combining agronomic advice with real-time price discovery and access to credit (Narendar et al., 2025; Singh, 2015) [35,52] . Finally, evidence from India and East Africa suggests that digitally enabled advisory services, when combined with quality data and localization, are associated with measurable gains in yield, input-use efficiency and market prices received by participating farmers, underscoring the economic case for continued investment in digital extension infrastructure (Gouroubera et [20,60,19] .

Challenges and Constraints

The opportunities associated with digital extension are counterbalanced by substantial and, in some respects, intensifying challenges. The most widely documented is the rural digital divide: many smallholder farmers, particularly in remote and underdeveloped regions, continue to lack reliable access to smartphones, mobile internet and stable electricity, constraining their ability to benefit from digital advisory platforms regardless of the sophistication of the underlying technology (Ragetlie et al., 2022; Sen et al., 2025) [41,50] . This divide is not merely infrastructural but socio-economic, correlating strongly with income, education and landholding size, so that digital extension risks disproportionately benefiting farmers who are already comparatively well resourced (Ahad et al., 2024; Sen et al., 2025) [3,50] . A closely related and persistent constraint is the gender digital divide. Women farmers frequently report lower rates of mobile phone and smartphone ownership, more limited decision-making autonomy over shared household devices, and lower levels of digital literacy than men, all of which constrain their independent access to digital advisory content even where household-level connectivity exists (Ragetlie et [41,33,3] . Addressing this divide requires extension systems to move beyond gender-neutral design toward deliberately gender-responsive digital tools and delivery strategies (Ragetlie et al., 2022) [41] . Digital literacy constitutes a further constraint that extends beyond farmers to extension personnel themselves. Studies from multiple regions report that the competencies of agricultural extension agents in using digital technologies remain uneven, undermining efforts to integrate digital tools into routine advisory practice and highlighting the need for sustained investment in extension-agent training (Deji et al., 2023; Angello, 2015) [12,7] . Content-related challenges are similarly significant: much digital advisory content has historically relied on generic recommendations poorly adapted to local languages, dialects, cropping systems and farming contexts, reducing perceived relevance and trust among target users (High et al., 2026; Sen et al., 2025) [24,50] . Data governance and privacy present emerging concerns as digital platforms increasingly collect farm-level, financial and even biometric data, raising questions about data ownership, consent, and the potential for commercial exploitation of farmer data by private technology providers without adequate regulatory oversight (Abate et al., 2023) [1] . The reliability and timeliness of information delivered through digital channels have also been questioned; assessments of established platforms such as Kisan Call Centres have found persistent gaps in the accuracy, currency and consistency of advisory content, despite generally high farmer satisfaction with accessibility (Tran et al., 2021) [57] . Furthermore, behavioural and trust-related barriersincluding skepticism toward AI-generated recommendations, preference for known human intermediaries, and limited farmer engagement with feedback mechanisms embedded in chatbot platformscontinue to constrain the effectiveness of even wellresourced digital advisory pilots (High et al., 2026) [24] . Finally, the sustainability of many digital extension initiatives remains uncertain beyond initial donor-or project-funded pilot phases, echoing the recurrent-cost problems that historically undermined the Training and Visit system, and underscoring the need for viable long-term financing and institutional ownership models (Rokvić & Vaško, 2016; Pandey et al., 2025) [45,40] .

Institutional Readiness and Policy Framework

Realising the potential of digital extension depends heavily on the readiness of public extension institutions, the coherence of national policy frameworks, and the extent to which digitalization is embedded within, rather than layered on top of, existing advisory structures. Institutional readiness encompasses several dimensions: the digital competency of extension personnel, the availability of interoperable data systems, the existence of quality-assured and regularly updated agronomic content, and organizational cultures that reward innovation and farmercentred service delivery (Regan, 2021; Deji et al., 2023) [42,12] . Assessments of publicly funded agricultural research and extension systems indicate that formal recognition of digital advisory work, together with dedicated funding streams and monitoring mechanisms, remains uneven across countries, constraining the translation of pilot successes into sustained institutional practice (Regan, 2021) [42] . Public-private partnerships (PPPs) have emerged as a central institutional mechanism for scaling digital extension, combining the technical and financial capacity of private technology providers, agribusinesses and telecommunications companies with the legitimacy, reach and agronomic knowledge base of public extension systems (Pandey et al., 2025; Moreddu, 2016) [40,30] . Comparative evidence from multiple countries indicates that PPPs bundling technology transfer, market linkage and capacitybuilding components consistently outperform singlecomponent interventions, though their success depends on clear role differentiation among partners, transparent governance arrangements, and equitable benefit-sharing mechanisms (Pandey et al., 2025; Aseete et al., 2023) [40,8] . Effective PPP design further requires standardised legal and contractual frameworks and extension-specific curricula capable of preparing personnel to operate within pluralistic, multi-stakeholder advisory ecosystems (Pandey et al., 2025) [40] . At the policy level, national digital agriculture strategiessuch as India's National Mission on Agricultural Extension and Technology and its more recent Digital Agriculture Mission, alongside similar frameworks emerging in parts of sub-Saharan Africa and Southeast Asia-seek to provide overarching direction for infrastructure investment, data governance and the harmonization of fragmented, often donor-driven digital pilots (Pandey et al., 2025; Regan, 2021) [40,42] . However, policy frameworks in many countries continue to lag behind technological deployment, with regulatory gaps particularly evident in relation to farmer data protection, quality assurance of AI-generated advisory content, interoperability standards across competing platforms, and mechanisms for accrediting private digital advisory providers operating within otherwise public extension mandates (Abate et al., 2023) [1] . Institutional readiness also requires explicit attention to rural infrastructure investment-electricity, broadband and mobile network coverage-as a precondition for the equitable functioning of any digital extension strategy, together with sustained public commitment to strengthening human capital and digital literacy among both farmers and extension personnel (Ragetlie et al., 2022; Sen et al., 2025) [41, 50] . Monitoring and evaluation systems capable of tracking not only the reach but the quality, relevance and equity of digital advisory delivery are similarly identified as a critical, and frequently underdeveloped, component of institutional readiness (Sen et al., 2025; Deji et al., 2023) [50,12] .

Case Studies and Comparative Analysis

A comparative examination of digital extension initiatives across countries illustrates both the transformative potential and the persistent limitations of digitalisation in practice.

India offers one of the most extensively documented experiences, spanning nearly two decades of digital extension innovation. The Kisan Call Centre, launched in January 2004, provides farmers with toll-free, locallanguage telephone access to agricultural experts and remains among the largest publicly operated digital advisory services in the world; evaluations report generally high farmer satisfaction with accessibility, although concerns persist regarding the accuracy and timeliness of advice delivered (Oumer, 2025; Singh, 2015; Tran et al., 2021) [38,52,57] . The privately operated e-Choupal network, established by ITC Limited in 2000, combined internet-enabled village kiosks with market-linkage and input-supply services, and has since evolved into a mobile-first platform integrating artificial intelligence and IoT-based agronomic advisory, illustrating a private-sector pathway to digital extension scaling, albeit one that has drawn scrutiny regarding uneven geographic coverage and dependence on a single corporate actor (Singh, 2015) [52] . In sub-Saharan Africa, mobile-based advisory services have been extensively piloted and evaluated, with meta-analytic evidence indicating generally positive, though heterogeneous, effects on technology adoption across multiple countries (Gouroubera et al., 2025; Muromba et al., 2025) [20,32] . Ethiopia's national farmer hotline demonstrates the potential of low-bandwidth, voice-based advisory services to reach farmers with limited smartphone access, though uptake remains constrained by low digital literacy and infrastructural gaps (Oumer, 2025) [38] . Ghana's postpandemic e-extension experience highlights both the accelerated adoption of ICT-based advisory channels during COVID-19 and the continuing need to integrate digital strategies with existing extension methods to ensure inclusive reach, particularly for marginalised communities (Siankwilimba et al., 2023) [51] . Gender-focused digital extension interventions, such as a household-targeted programme in Tunisia that engaged husband-and-wife farmer pairs, provide encouraging evidence that deliberately gender-responsive design can narrow, rather than reproduce, gender gaps in digital advisory access (Ragetlie et al., 2022) [41] . Emerging generative AI-based advisory platforms, including Digital Green's Farmer.Chat, deployed in partnership with national agriculture ministries in East Africa and piloted in South Asia, represent the newest generation of digital extension and combine curated, government-vetted knowledge bases with conversational large language model interfaces capable of responding to open-ended farmer queries in local languages (High et al., 2026; Feyisa, 2020) [24, 16] . Early evidence suggests strong potential for scaling personalised advisory delivery, though systematic evaluation of yield, income and behavioural impacts remains limited, and challenges relating to content accuracy, farmer trust and feedback-loop engagement persist (High et al., 2026) [24] . A comparative summary of these initiatives is presented in Table 3. [38,52,57] e-Choupal, India Internet kiosks / mobile app (private sector)

Integrated market linkage and input supply

Uneven coverage; single-firm dependence (Singh, 2015) [52] Farmer's Hotline, Ethiopia Mobile voice advisory (short-code dial-in)

Low-bandwidth accessibility

Low uptake; limited digital literacy (Oumer, 2025) [38] Gender-responsive digital extension, Tunisia [54,50,49] .

Future Research Directions

Several priority areas warrant further research attention as digital extension systems mature. First, rigorous, longerterm impact evaluations-particularly randomised and quasiexperimental studies tracking yield, income, resilience and empowerment outcomes rather than adoption or usage metrics alone-remain scarce relative to the proliferation of digital extension pilots, and are needed to establish a credible evidence base for scaling decisions (Zheng & Ma, 2023; Giulivi et al., 2023) [60,19] . Second, research is needed on the design and evaluation of generative AI and large language model-based advisory systems specifically within smallholder contexts, including their accuracy, bias, explainability, and the mechanisms through which farmer trust in AI-generated recommendations is built or eroded over time (Neeraja et al., 2026; High et al., 2026) [37,24] . Third, further work is required on gender-and equitysensitive design of digital extension platforms, including how household dynamics, intra-household device sharing, and social norms mediate women's and youth's actual use of nominally available digital tools, moving beyond access metrics toward a deeper understanding of effective use and empowerment (Mutuku et al., 2026; Ahad et al., 2024) [33,3] . Fourth, comparative institutional research on public-private partnership governance models-examining risk-sharing, data ownership, accountability and long-term financial sustainability-would help clarify which partnership configurations are best suited to different country contexts and extension mandates (Pandey et al., 2025; Aseete et al., 2023) [40,8] . Fifth, the integration of IoT, remote sensing and blockchain technologies into mainstream extension advisory workflows, as opposed to isolated technical pilots, remains underexplored, particularly regarding the interoperability of these data streams with the knowledge bases underpinning farmer-facing advisory chatbots (Safeer et [47,17,14] . Sixth, research on data governance frameworks appropriate to agricultural extension contexts-covering farmer consent, data portability and protection against exploitative commercial use of farmlevel data-is needed to keep pace with the rapid expansion of data-intensive digital advisory platforms (Abate et al., 2023) [1] . Finally, systematic research on the digital competencies, incentive structures and career pathways of extension personnel operating within increasingly hybrid human-digital advisory systems would help ensure that the human extension workforce is strengthened, rather than displaced, by digitalisation (Angello, 2015; Deji et al., 2023) [7,12] .

Conclusion

Agricultural extension has evolved from a conventional topdown technology transfer approach to a more participatory and digitally enabled advisory system. While traditional extension has played a vital role in improving agricultural productivity, its effectiveness has been constrained by inadequate extension personnel, limited financial resources, weak research-extension linkages, and challenges in reaching diverse farming communities. Recent advances in digital technologies including mobile applications, webbased platforms, remote sensing, the Internet of Things (IoT), blockchain, and artificial intelligence (AI) have significantly expanded the capacity of extension systems to deliver timely, location-specific, and personalized advisory services. Evidence reviewed in this study indicates that these technologies have improved farmers' access to information, enhanced technology adoption, and strengthened agricultural decision-making across many regions. Despite these opportunities, several challenges continue to hinder the widespread adoption of digital extension. Persistent digital divides, gender disparities, inadequate infrastructure, limited digital literacy, concerns regarding data privacy and governance, and the long-term sustainability of digital initiatives remain significant barriers. The effectiveness of digital extension therefore depends not only on technological innovation but also on supportive policies, institutional capacity, trained extension professionals, and effective public-private partnerships. Overall, the evidence suggests that the future of agricultural extension lies in hybrid advisory systems that integrate the scalability and efficiency of digital technologies with the contextual knowledge, trust, and interpersonal support provided by human extension agents. Future research should emphasize rigorous impact evaluation, inclusive digital innovation, ethical AI applications, and stronger institutional frameworks to ensure that digital transformation contributes to equitable, resilient, and sustainable agricultural development. Such an integrated approach will be essential for delivering effective extension services capable of addressing the evolving challenges of global agriculture.

References

1. Abate GT, Abay KA, Chamberlin J, Kassim Y, Spielman DJ, Tabe-Ojong MP Jr. Digital tools and agricultural market transformation in Africa: Why are they not at scale yet, and what will it take to get there? Food Policy. 2023;116:102439. https://doi.org/10.1016/j.foodpol.2023.102439
2. Adesiji GB, Adelowo JY, Komolafe SE, Adesiji TT. Farmers' perceived rating and usability attributes of agricultural mobile phone apps. Smart Agric Technol. 2024;8:100501. https://doi.org/10.1016/j.atech.2024.100501
3. Ahad B, Ahmad S, Iqbal M. Gender exclusiveness of agricultural technology and IoT (Internet of Things) and its impact on agricultural extension services. Educ Admin Theory Pract. 2024;30(5):13876-13879. https://doi.org/10.53555/kuey.v30i5.6104
4. Alipio M, Villena ML. Intelligent wearable devices and biosensors for monitoring cattle health conditions: A review and classification. Smart Health. 2023;27:100369. https://doi.org/10.1016/j.smhl.2022.100369
5. Al-Kaisi MM, Elmore RW, Guzman JG, Hanna HM, Hart CE, Helmers MJ, et al. Extension agriculture and natural resources in the U.S. Midwest: A review and analysis of challenges and future opportunities. Nat Sci Educ. 2015;44(1):26-33. https://doi.org/10.4195/nse2014.10.0022
6. Almazmomi NK. Artificial intelligence-driven blockchain and Internet of Things framework for secure data management in precision agriculture. Meas Control. 2025. https://doi.org/10.1177/09266801251331816
7. Angello C. Potential of information and communication technologies in promoting access to livestock information: Perceptions of urban livestock keepers in Tanzania. Int J Inf Commun Technol Hum Dev. 2015;7(2):20-41. https://doi.org/10.4018/IJICTHD.2015040102
8. Aseete P, Barkley A, Katungi E, Ugen MA, Birachi E. Public-private partnership generates economic benefits to smallholder bean growers in Uganda. Food Secur. 2023;15(1):201-218. https://doi.org/10.1007/s12571022-01309-5
9. Chao J, Li TS, Yin HL, Wang Z. Adoption of multiple sustainable agricultural practices among farmers in northwest China. Cogent Food Agric. 2024;10(1):2300189. https://doi.org/10.1080/23311932.2023.2300189
10. Chen HY, Sharma K, Sharma C, Sharma S. Integrating explainable artificial intelligence and blockchain to smart agriculture: Research prospects for decision making and improved security. Smart Agric Technol. 2023;6:100350. https://doi.org/10.1016/j.atech.2023.100350
11. Cheng CM, Gao Q, Qiu YQ. Assessing the ability of agricultural socialized services to promote the protection of cultivated land among farmers. Land. 2022;11(8):1338. https://doi.org/10.3390/land11081338
12. Deji OF, Famakinwa M, Alabi DL, Faniyi EO. Utilisation of artificial intelligence-based technology for agricultural extension services among extension professionals in Nigeria. J Agric Ext. 2023;27(3):80-90. https://doi.org/10.4314/jae.v27i3.9
13. Delgoda D, Malano H, Saleem SK, Halgamuge MN. Irrigation control based on model predictive control (MPC): Formulation of theory and validation using weather forecast data and AQUACROP model. Environ Model Softw. 2016;78:40-53. https://doi.org/10.1016/j.envsoft.2015.12.012
14. Dey K, Shekhawat U. Blockchain for sustainable eagriculture: Literature review, architecture for data management, and implications. J Clean Prod. 2021;316:128254. https://doi.org/10.1016/j.jclepro.2021.128254
15. Feder G, Slade RH, Lau LJ. Does agricultural extension pay? The training and visit system in Northwest India. Am J Agric Econ. 1987;69(3):677-686. https://doi.org/10.2307/1241702
16. Feyisa BW. Determinants of agricultural technology adoption in Ethiopia: A meta-analysis. Cogent Food Agric. 2020;6:1855817. https://doi.org/10.1080/23311932.2020.1855817
17. Friha O, Ferrag MA, Shu L, Maglaras L, Wang X. Internet of Things for the future of smart agriculture: A comprehensive survey of emerging technologies. IEEE/CAA J Autom Sin. 2021;8(4):718-752. https://doi.org/10.1109/jas.2021.1003925
18. Gao T, Lu Q, Zhang Y, Feng H. The influence of digital agricultural technology extension service cognition on farmers' pro-environmental behaviour: The moderating role of policy support. Int J Agric Sustain. 2025;23(1). https://doi.org/10.1080/14735903.2025.2569186
19. Giulivi N, Harou AP, Gautam S, Guereña D. Getting the message out: Information and communication technologies and agricultural extension. Am J Agric Econ. 2023;105:1011-1045. https://doi.org/10.1111/ajae.12348
20. Gouroubera MW, Adechian SA, Segnon AC, Moumouni-Moussa I, Zougmoré RB. Drivers and impacts of mobile phone-mediated scaling of agricultural technologies: A meta-analysis. Front Sustain Food Syst. 2025;8:1514546. https://doi.org/10.3389/fsufs.2024.1514546
21. Gouroubera MW, Kora Sabi A, Bio Comada TK, Dosso F, Fatondji SA, Gouthon MB. Designing effective digital-based delivery of climate information for smallholder farmers: A mini meta-analysis on drivers and barriers. Clim Policy. 2023;24:1443-1456. https://doi.org/10.1080/14693062.2023.2266475
22. Gupta G, Pal SK. Applications of AI in precision agriculture. Discover Agric. 2025;3:61. https://doi.org/10.1007/s44279-025-00220-9
23. Hassoun A, Marvin HJP, Bouzembrak Y, Barba FJ, Castagnini JM, Pallarés N, et al. Digital transformation in the agri-food industry: Recent applications and the role of the COVID-19 pandemic. Front Sustain Food Syst. 2023;7:1217813. https://doi.org/10.3389/fsufs.2023.1217813
24. High C, Singh N, Nemes G. Artificial intelligence for agricultural extension: Supporting transformative learning among smallholder farmers. J Ext Educ. 2026. https://doi.org/10.1177/24551333251345224
25. Kaushik I, Prakash N, Jain A. An AI-blockchainassisted smart agriculture framework for enabling secure and efficient data transaction: A hybrid approach. Knowl Inf Syst. 2025;67:10087-10135. https://doi.org/10.1007/s10115-025-02526-y
26. Khan NA, Qijie G, Sertse SF, Nabi MN, Khan P. Farmers' use of mobile phone-based farm advisory services in Punjab, Pakistan. Inf Dev. 2020;36(3):390402. https://doi.org/10.1177/0266666919864126
27. Khan RP, Gupta S, Daum T, Birner R, Ringler C. Levelling the field: A review of the ICT revolution and agricultural extension in the Global South. J Int Dev. 2025;37(1):1-21. https://doi.org/10.1002/jid.3949
28. Malik YS, Ansari MI, Gharieb R, Ghosh S, Chaudhary RK, Hemida MG, et al. The impact of COVID-19 pandemic on agricultural, livestock, poultry and fish sectors. Int J Vet Sci Med. 2024. https://doi.org/10.1155/2024/5540056
29. Mane D, Sangve S, Halmadge A, Takale A, Choubhare N, Shinde A. Smart digital e-portal for Indian farmers. In: Inventive communication and computational technologies. ICICCT 2023 (Lecture Notes in Networks and Systems, Vol. 757). Cham: Springer; 2023. https://doi.org/10.1007/978-981-99-5166-6_75
30. Moreddu C. Public-private partnerships for agricultural innovation: Lessons from recent experiences. OECD Food, Agriculture and Fisheries Papers, No. 92. Paris: OECD Publishing; 2016. https://doi.org/10.1787/5jm55j9p9rmx-en
31. Mulungu K, Kassie M, Tschopp M. The role of information and communication technologies-based extension in agriculture: Application, opportunities and challenges. Inf Technol Dev. 2025;31(4):1117-1146. https://doi.org/10.1080/02681102.2025.2456232
32. Muromba P, Keeni M, Fuyuki K. A systematic review of mobile agricultural service applications for smallholder farmers in sub-Saharan Africa: Perspectives from the technology acceptance model. Agric Food Secur. 2025;14:34. https://doi.org/10.1186/s40066-025-00563-y
33. Mutuku B, Akuffobea-Essilfie M, Kansiime M, Duah S, Oppong-Mensah B, Mibei H, et al. Understanding gender disparities in access to digital advisory services among women and men farmers in Ghana. Front Sustain Food Syst. 2026;10:1614245. https://doi.org/10.3389/fsufs.2026.1614245
34. Narendar G, Narendar J, Dash RB, Kar I, Baliarsingh A, Parameswari YS. Technological advances in weed management for sustainable crop production. 2026;57(01):21897-21908.
35. Narendar G, Rakesh G, Saicharan M, Dinesh A, Krishna Chaitanya A, Balaji Naik B. Different constraints affecting sugarcane production and suggestions for improvement of sugarcane area and production in Telangana state. AATCC J. 2025;13(02):485-488. https://doi.org/10.21276/AATCCReview.2025.13.02.48 5
36. Ndimbo GK, Yu L, Buma AAN. ICTs, smallholder agriculture and farmers' livelihood improvement in developing countries: Evidence from Tanzania. Inf Dev. 2025. https://doi.org/10.1177/02666669231165272
37. Neeraja B, Suryawanshi D, Taye N, Kumar R, Bharti S, Tripathi S. Artificial intelligence in agricultural extension: Transforming farmer advisory systems, decision support and knowledge delivery. J Exp Agric Int. 2026;48(6):352-371. https://doi.org/10.9734/jeai/2026/v48i64288
38. Oumer M. Determinants affecting smallholder farmers' use of mobile-based agricultural information via the Farmer's Hotline in Southern Ethiopia's Borecha District. Adv Agric. 2025. https://doi.org/10.1155/aia/4157664
39. Pal A, Kant K. Using blockchain for provenance and traceability in Internet of Things-integrated food logistics. Computer. 2019;52(12):94-98. https://doi.org/10.1109/mc.2019.2942111
40. Pandey MK, Singh M, Sulekha, Rana P, Tripathi S, Vishwakarma SK, et al. Public-private partnerships in agricultural extension: Opportunities and challenges. J Exp Agric Int. 2025;47(6):597-612. https://doi.org/10.9734/jeai/2025/v47i63520
41. Ragetlie R, Najjar D, Oueslati D. "Dear brother farmer": Gender-responsive digital extension in Tunisia during the COVID-19 pandemic. Sustainability. 2022;14(7):4162. https://doi.org/10.3390/su14074162
42. Regan Á. Exploring the readiness of publicly funded researchers to practice responsible research and innovation in digital agriculture. J Resp Innov. 2021;8(1):28-47. https://doi.org/10.1080/23299460.2021.1904755
43. Rehman KU, Andleeb S, Ashfaq M, Akram N, Akram MW. Blockchain-enabled smart agriculture: Enhancing data-driven decision making and ensuring food security. J Clean Prod. 2023;427:138900. https://doi.org/10.1016/j.jclepro.2023.138900
44. Rogers EM. The intellectual foundation and history of the agricultural extension model. Knowledge. 1988;9(4):492-510. https://doi.org/10.1177/0164025988009004003
45. Rokvić G, Vaško Ž. Stages of development of agricultural extension service in Bosnia and Herzegovina. Agroznanje. 2016;17(4):359-369. https://doi.org/10.7251/AGREN1604359R
46. Roy S, Singh S, Yogesh BM, Vyas M, Shekhar G, Mukherjee S, et al. Generative AI and chatbot-based advisory: The next frontier in agricultural extension delivery. Int J Agric Sci. 2026;9(2Sa). https://doi.org/10.33545/2618060X.2026.v9.i2Sa.4881
47. Safeer S, Mastro GD, Pulvento C. IoT based climate smart agriculture succeeded by blockchain database-A bibliometric analysis. Front Sustain Food Syst. 2024;8:1406871. https://doi.org/10.3389/fsufs.2024.1406871
48. Sai Satish D, Narendar G, Mounika T, Prameela G. Innovations in agriculture-smart solutions for sustainable food production. Int J Res Agron. 2025;8(6):1028-1033. https://doi.org/10.33545/2618060X.2025.v8.i6m.3145
49. Sen LTH, Chou P, Dacuyan FB, Nyberg Y, Wetterlind J. Barriers and enablers of digital extension services' adoption among smallholder farmers: The case of Cambodia, the Philippines and Vietnam. Int J Agric Sustain. 2024;22(1). https://doi.org/10.1080/14735903.2024.2368351
50. Sen LTH, Phuong LTH, Chou P, Dacuyan FB, Nyberg Y, Wetterlind J. The opportunities and barriers in developing interactive digital extension services for smallholder farmers as a pathway to sustainable agriculture: A systematic review. Sustainability. 2025;17(7):3007. https://doi.org/10.3390/su17073007
51. Siankwilimba E, Hiddlestone-Mumford J, Hoque ME, Hang'ombe BM, Mumba C, Hasimuna OJ, et al. Sustainability of agriculture extension services in the face of COVID-19: A study on gender-specific market systems. Cogent Food Agric. 2023;9(2):2284231. https://doi.org/10.1080/23311932.2023.2284231
52. Singh H. The impact of Kisan Call Centers on the farming sector. In: Promoting socio-economic development through business integration. Hershey: IGI Global; 2015. https://doi.org/10.4018/978-1-4666-82597.ch006
53. Singh SS, Singh RJ, Devarani L, Hemochandra L, Singh R, Choudhury A. COVID-19 post-pandemic agricultural extension services: A strategic analysis of post-pandemic sustainability. Arch Curr Res Int. 2025;25(7):517-536. https://doi.org/10.9734/acri/2025/v25i71355
54. Steinke J, van Etten J, Müller A, Ortiz-Crespo B, van de Gevel J, Silvestri S, et al. Tapping the full potential of the digital revolution for agricultural extension: An emerging innovation agenda. Int J Agric Sustain. 2021;19(5-6):549-565. https://doi.org/10.1080/14735903.2020.1738754
55. Subeesh A, Mehta CR. Automation and digitization of agriculture using artificial intelligence and Internet of Things. Artif Intell Agric. 2021;5:278-291. https://doi.org/10.1016/j.aiia.2021.11.004
56. Tang A, Tchao ET, Agbemenu AS, Keelson E, Klogo GS, Kponyo JJ. Assessing blockchain and IoT technologies for agricultural food supply chains in Africa: A feasibility analysis. Heliyon. 2024;10(15):e34584. https://doi.org/10.1016/j.heliyon.2024.e34584
57. Tran QN, Turnbull BP, Wu HT, de Silva AJS, Kormusheva K, Hu J. A survey on privacy-preserving blockchain systems (PPBS) and a novel PPBS-based framework for smart agriculture. IEEE Open J Comput Soc. 2021;2:72-84. https://doi.org/10.1109/OJCS.2021.3053032
58. Yang Y, Lin M, Lin Y, Zhang C, Wu C. A survey of blockchain applications for management in agriculture and livestock Internet of Things. Future Internet. 2025;17(1):40. https://doi.org/10.3390/fi17010040
59. Yegbemey RN, Egah J. Reaching out to smallholder farmers in developing countries with climate services: A literature review of current information delivery channels. Clim Serv. 2021;23:100253. https://doi.org/10.1016/j.cliser.2021.100253
60. Zheng H, Ma W. Smartphone-based information acquisition and wheat farm performance: Insights from a doubly robust IPWRA estimator. Electron Commer Res. 2023;23:633-658. https://doi.org/10.1007/s10660021-09481-0