Knowledge, Innovation and Technology in Scientific Research Vol. 6

Extension 5.0: Human-Centred Digitalisation, Artificial Intelligence and the Future of Agricultural Advisory Services

2026-10-28T10:12:45+00:00

Agricultural advisory services are being reshaped by digital platforms, data-intensive decision support and, increasingly, artificial intelligence (AI). Yet the dominant progression from conventional extension to digital extension and Agriculture 4.0 has often treated scale, automation and information throughput as proxies for effective advisory performance. This critical narrative review examines whether a more human-centred paradigm is warranted and develops "Extension 5.0" as a provisional analytical concept for advisory systems that combine computational capability with human judgement, social learning, inclusion and accountable governance. Literature published from 1 January 2000 to 7 August 2026 was identified through multidisciplinary and agriculture- and technology-oriented scholarly sources, supplemented by citation searching and DOI-level bibliographic verification. Evidence was synthesised across agricultural economics, extension studies, rural sociology, innovation studies, human-computer interaction and responsible AI. Digital advisory interventions can improve knowledge, recommended-practice adoption and, in some settings, farm performance, but effects are heterogeneous and do not consistently progress from information access to durable welfare gains. Personalisation improves relevance, yet requires reliable local data and does not remove price, credit, labour, infrastructure or risk constraints. Emerging generative AI can lower the cost of translating complex knowledge into dialogue, but hallucination, uncertainty, bias, weak localisation and accountability make unsupervised agronomic advice unsafe as a default institutional model. The strongest case is therefore for hybrid systems in which AI augments advisers and farmers, escalation paths remain available, and design is organised around capability, trust, data rights and contextual fit. Gender, age, literacy, language and resource inequalities are not peripheral adoption variables; they shape who can exercise agency within digital systems. Extension 5.0 is consequently best understood not as a settled technological stage but as a socio-technical design proposition. Its value will depend on evidence that hybrid human-AI systems improve decision quality, equity, resilience and farmer autonomy rather than merely increasing digital contact or automation.

AI Technology: Reshaping the Pedagogical Practices

2026-10-28T10:18:56+00:00

Artificial intelligence has extended educational automation beyond restricted tutoring and prediction systems into everyday activities of lesson preparation, dialogue, feedback and assessment. Yet the capacity to produce useful instructional outputs does not establish that learners acquire durable understanding. This critical narrative review examines how adaptive, predictive and generative systems redistribute cognitive work and pedagogical judgement across students, teachers and institutions. Relevant peer-reviewed literature and authoritative educational guidance were selected through topic-specific scholarly searches, source verification and thematic appraisal, with particular attention to comparative instructional studies, recent evidence syntheses and ethical scholarship. The most established quantitative evidence concerns structured intelligent tutoring, although effects depend on curricular fit, comparator and assessment. Emerging randomised studies of generative tutoring reveal opposing outcomes: carefully designed scaffolding can improve short-term learning under specific conditions, whereas relatively unguarded assistance may improve performance during use while weakening subsequent unaided achievement. Evidence on automated feedback similarly requires distinctions among comment quality, perceived usefulness, revision uptake and transfer. Generative systems also unsettle the validity of conventional assessment products and alter professional responsibilities for explanation, verification and learner support. Promising uses of AI literacy instruction, human-centred analytics and hybrid teacher–AI orchestration remain constrained by differences in access, local context, data governance and limited longitudinal evaluation. The review proposes an integrated appraisal centred on task design, system mediation, teacher judgement, learner activity and demonstrable independent competence. The educational value of AI is therefore conditional, not technologically predetermined. Future research should compare credible alternatives, study mechanisms and delayed outcomes, and test whether purported benefits persist across diverse schools, disciplines and learner groups.

Responsible Artificial Intelligence in Agricultural Extension: Accuracy, Accountability, Bias, Privacy and Farmer Protection

2026-10-28T10:28:01+00:00

Artificial intelligence (AI), particularly generative AI and large language models, is entering agricultural extension at a point when advisory systems are already under pressure to provide timely, locally specific knowledge at scale. This review critically examines responsible AI in agricultural extension through five connected domains: accuracy, accountability, bias, privacy and farmer protection. A critical narrative approach was used to integrate evidence from agricultural extension, digital agriculture, AI ethics, data governance and recent evaluations of AI-enabled advisory tools. The literature indicates that AI can widen access to information, assist extension personnel and support localisation of technical content, but the evidence base remains stronger for feasibility and expert-rated response quality than for sustained farmer-level outcomes. Accuracy is context dependent: fluent answers can remain agronomically unsafe when planting dates, input rates, disease-control options or local regulations are wrong or outdated. Accountability is frequently diffuse across foundation-model providers, application developers, knowledge curators, extension organisations and human advisers. Bias can arise from unequal connectivity, under-represented languages and agroecologies, gendered assumptions, farm-size effects and commercial incentives, while aggregate accuracy can conceal unequal harms. Privacy concerns extend beyond conventional personal data to economically sensitive farm, location, production and behavioural information whose value may emerge through inference or linkage. The review argues that farmer protection should be treated as an operational safety-and-governance obligation rather than as a disclosure exercise. Responsible deployment therefore requires locally grounded evidence, risk-tiered human oversight, subgroup testing, data minimisation, traceability, incident management, accessible complaints and meaningful remedies. A proposed extension-system governance model links evidence governance, model governance, deployment safeguards, institutional accountability and farmer-facing protections. The principal research need is independent, longitudinal and multi-context evaluation that measures not only technical performance but also safety incidents, distributional effects, trust, autonomy, economic outcomes and the quality of human-AI extension relationships.