Aug 2026· Big Data and Cognitive Computing· 0 citations· 111 references
TL;DR
A five-dimensional diagnostic framework that maps the challenges of human-AI collaboration across Integration, Representation, Scale, Temporality, and Adequacy gaps and shows that augmentation remains the dominant and most viable mode of use in complex environments.
Abstract
The growing capabilities of artificial intelligence (AI) have not translated straightforwardly into organisational value. A persistent disconnect—the “last-mile problem”—arises from structural gaps between idealised AI tasks and real-world organisational contexts. Synthesising insights from organisational theory, cognitive science, and computer science, we have developed a five-dimensional diagnostic framework that maps the challenges of human-AI collaboration across Integration, Representation, Scale, Temporality, and Adequacy gaps. These gaps illuminate how socio-technical complexity, contextualised problem representations, interdependencies among agents, dynamic environments, and limitations in current AI reasoning collectively constrain full automation and demand human judgement. By reviewing the historical evolution of AI—from symbolic systems to machine learning, generative models, and emerging agentic approaches—we show that augmentation remains the dominant and most viable mode of use in complex environments. An illustrative system-dynamics example demonstrates how improvements in algorithmic performance do not automatically yield proportional system-level gains. Overall, our framework provides researchers with a conceptual lens and practitioners with a diagnostic tool for assessing complementarities and informing the design of human-AI collaborations. The framework is offered as a conceptual synthesis and diagnostic instrument rather than an empirically validated model.
Artificial Intelligence (AI) enables powerful capabilities that are transforming almost all sectors. However, the economic growth driven by AI comes at a cost, and its sociotechnical impacts are fraught with contradictions and paradoxes. As a result, several legal initiatives and risk management frameworks have been introduced to mitigate the various risks associated with AI systems. Agentic AI systems require even closer attention than traditional AI. While traditional AI has a narrow focus and responds to direct commands, Agentic AI emerges from combining multiple types of AI capable of planning, tool use, and multi-step execution. These systems can behave and interact autonomously, making decisions and performing tasks to achieve system objectives with minimal human oversight. Recognizing that Agentic AI represents a paradigm shift, this paper addresses its challenges from a Human-AI Interaction perspective. It examines the root causes and impacts of risks arising from the transition from Task Automation to Intentionality Automation, where the user manages outcomes and constraints rather than individual task steps. Key issues include the Open-Loop Control Gap and the Metacognitive Gap, whose relationship is fundamental to understanding the collapse of human oversight, as they represent two sides of the same coin in the loss of control. By analysing scenarios such as cybersecurity and healthcare, this paper identifies dimensions of user demand and identifies Ecological Interface Design as an ergonomic approach to ensure that as AI gains agency, the human retains authority and situational awareness.
M. Simões-Marques· AHFE International· 0 citations
The integration of Artificial Intelligence (AI) as a collaborative partner is transforming the future of work. Rather than replacing human labor, modern AI systems enhance human capabilities through cognitive augmentation, adaptive workflows, and cooperative problem-solving. This paper presents a multidisciplinary analysis of human–AI collaboration across sectors such as healthcare, engineering, finance, education, and creative industries. A conceptual framework is proposed for dynamic task allocation between humans and AI based on uncertainty, contextual reasoning, and interpretability requirements. The study also examines socio-technical challenges including trust, ethical alignment, skill transformation, and organizational resilience. Using a domain-agnostic evaluation approach, collaboration effectiveness is measured through metrics such as cognitive load distribution, error reduction, adaptability, and explainability. The findings indicate that hybrid intelligence systems outperform both purely human and fully automated systems in complex and uncertain environments. The study concludes that the future of work will depend on co-evolutionary human-AI collaboration, requiring organizational restructuring, policy development, and ethical safeguards to ensure sustainable productivity and innovation.
V. Sethi· International Journal of Eme...· 0 citations
The accelerating diffusion of artificial intelligence (AI) across economic sectors has fundamentally altered the nature of work, organisational structures, and global labour ecosystems, generating intense scholarly and policy debates regarding the future relationship between humans and intelligent technologies. Prevailing discourse has largely been shaped by automation-centric and technologically deterministic perspectives that portray AI as an inevitable substitute for human labour, often overlooking the growing evidence that enduring organisational performance increasingly depends on synergistic interactions between human expertise and machine intelligence. This paper reconceptualises workforce transformation by positioning human complementarity as the central mechanism through which AI creates sustainable economic and organisational value. It develops an integrative conceptual framework that explains how AI enhances, rather than replaces, human capabilities in strategic reasoning, contextual judgement, creativity, ethical deliberation, adaptive learning, and collaborative decision-making across diverse global industries. The study further examines the institutional, technological, educational, and governance enablers required to cultivate effective human-AI partnerships while addressing emerging challenges associated with workforce restructuring, digital inequality, algorithmic bias, and evolving competency requirements. By shifting analytical attention from technological substitution to capability augmentation and co-evolution, the paper advances a human-centred interpretation of AI-enabled workforce transformation that enriches theoretical understanding and provides strategic insights for organisations, policymakers, educational institutions, and labour market stakeholders navigating the transition towards intelligent and inclusive work systems.
Raymond Ayivor· International Journal of Res...· 0 citations
As AI systems become increasingly integrated into consequential domains such as healthcare, journalism, education, scientific research, organizational decision-making, and defense, effective human-AI collaboration has emerged as a critical challenge. However, the sociotechnical risks that undermine collaboration are often studied in isolation, obscuring the recurring failure mechanisms that cut across domains. This paper presents a lifecycle-oriented synthesis of human-AI collaboration risks spanning four stages: task allocation, interaction, feedback, and adoption. Drawing on evidence from diverse application domains, we identify six recurring cross-domain risk clusters: Trust Miscalibration, Cognitive Burden, Accountability Gap, Capability Erosion, Goal Misalignment, and AI Anxiety and Technostress. We further propose a conceptual interaction model that illustrates how these risks emerge from sociotechnical drivers, interact through cascading pathways, and ultimately affect team performance and human well-being. Our analysis shows that many collaboration failures stem not from isolated technical deficiencies but from interconnected sociotechnical dynamics, helping explain why piecemeal interventions frequently create unintended consequences. By synthesizing fragmented literature into a unified framework, this work provides a foundation for future empirical research, lifecycle-oriented governance, and the design of more resilient, trustworthy, and human-centered human-AI collaboration systems.
Md Foysal Ahmed, Isaac Kobby Anni, Md Main Uddin Rony· 0 citations
Large Language Models (LLMs) are increasingly integrated into agentic workflows that require extended reasoning, persistent state management, coordinated tool use, and controlled execution. As this operational scope expands, a central question emerges: whether probabilistic generation alone can reliably support coherent behavior across interacting system components. This paper addresses that question through a structural diagnostic review of contemporary agentic systems. Starting from LLM-based tutoring as an analytically demanding entry point and extending toward structurally related agent architectures, the paper draws on a five-phase review of N=145 research records. The analysis is organized through the Agentic Structure Taxonomy (AST), which structures the literature across four dimensions: Cognition, Interaction, Orchestration, and Governance. The review identifies five recurrent empirical problem patterns and uses them as abductive diagnostic cues for formulating seven cross-dimensional transition gaps that capture recurrent discontinuities at the boundaries between reasoning, state, control, and execution. From these gaps, fourteen structural constraints are derived across three control domains: state isolation, control alignment, and execution governance. These constraints are interpreted not as prescriptive design mandates, but as analytically derived conditions associated with reducing error propagation across subsystem transitions. The paper argues that reliability in agentic systems is shaped not only by model performance or prompt design, but also by whether the boundaries linking probabilistic reasoning to persistent state, orchestration, and execution are governed by explicit structural conditions.
Christopher Valdez-Cantú, J. A. Cantoral-Ceballos, Joanna Alvarado-Uribe· Applied Informatics· 0 citations
The integration of artificial intelligence (AI) into Information Systems (IS) research is driving unprecedented individual productivity while introducing systemic strains: methodological homogenization, workflow opacity, and citation polarization. We argue these pathologies are not transient technological glitches but symptoms of an epistemic fairness paradox: the AI capabilities that maximize fluent, high-volume throughput strain the methodological pluralism and contextual rigor required to study sociotechnical phenomena. Drawing upon the FAIR design theory [13], we translate the architecture of organizational AI fairness to the decentralized epistemic ecosystem and conceptualize the challenge as a single paradox spanning three dimensions of tension (principles, goals, and foci) and three coupled stakeholders: the researcher, the intermediary, and the ecosystem. Because the paradox is endogenous to a rapidly evolving, stochastic, and increasingly agentic technology, static policy will fail. We offer seven provocations, one adaptive cycle per stakeholder region of the paradox, designed to embed the continuous surfacing and provisional resolution of epistemic friction into the field’s core institutions, so that AI serves as an engine for pluralistic discovery rather than a homogenizing force.
Arun Rai· ACM Transactions on Manageme...· 0 citations