The findings suggest that Expertise-Building serves as a sufficient proxy for predicting user behavior, as the action cannot be completed without the essential Expertise-Building "gatekeeper", and contributes to both scholarly discourse and industry practice.
Abstract
The rapid advancement of near human Generative Artificial Intelligence (GAI) systems is reshaping the modern software engineering and development ecosystem, altering practitioners' workflows, cognitive processes, and technology engagement patterns. This research undertaking, through an empirical quantitative study, analyzed the structural drivers of Intended Use (IU) within a complex cognitive framework, challenging traditional technology acceptance paradigms that prioritize ease of use. Motivated by an industry rooted business problem of adoption variability in the fast-paced markets, competition and corporate dynamics, this study investigated the mechanisms through which software engineering and development practitioners build Expertise in GAI utilization and form Intentions to adopt GAI for algorithmic design and code generation and broader software engineering tasks. Thus, this study highlights the internalized logic--the mental model--that a developer builds when interacting with an AI. Grounded in extant theory and contextualized within business software engineering and development dynamics, the study examined how Familiarity, Trust, Perceived Usability, Perceived Behavioral Control, Deterrence, and Complexity influence GAI Expertise Building and Adoption decisions in the context of modern software engineering ecosystem. Therefore, this study intended to derive new findings and conclusions on what theory-based and industry factors contextualize and affect this process, with emphasis on the said ecosystem. The results revealed a highly efficient, parsimonious "engine" model. The negligible impact of deterrence (R²= 0.03) further underscored that user intent is driven by capability-based "pull" rather than an inhibitor-based "push". On a more granulated outlook, the major takeaway from this study is that although the predictors in our model would intuitively make sense individually, what I tried to unravel was to tell the story of the relative importance of each expected predictor within a parsimonious model, as well as the role of statistical controls in moderating these predictors. All predictors were anticipated to be important because we already know they affect the acceptance and adoption of other types of information technologies. The findings suggest that Expertise-Building serves as a sufficient proxy for predicting user behavior, as the action cannot be completed without the essential Expertise-Building "gatekeeper". GLM results in particular, emphasized that the "action" is a function of capability, not logistics. On an exploratory level, I also tested possible interaction among two important predictors, extending understanding of how adoption behaviors emerge in complex high-tech environments. Additional important implications, as well as study limitations and future research recommendations are also discussed. To enhance translational relevance, this study also addresses the practical challenges software engineering and development practitioners face during early-stage GAI adoption--particularly in enterprise grade software engineering contexts where GAI integration requires new models of learning, skill acquisition, trust calibration, and workflow adaptation. The findings offer actionable guidance and recommendations for developing corporate and organizational dynamic capabilities, such as GAI related learning curves, confidence in the utilization of GAI, and expertise development pathways. These insights support software engineers, development managers, R&D units, program leaders, and executive decision makers in designing GAI enabled workflows and methodologies, governance structures, and project strategies that enhance productivity, efficiency, and innovation across software manufacturing workflows. Collectively, the study contributes to both scholarly discourse and industry practice by illuminating how Expertise-Building in GAI technologies is cultivated and how such Expertise drives effective and confident Adoption in the software engineering and development ecosystem. Keywords: expertise-building of GAI utilization, GAI adoption for algorithm design, software engineering with AI, GAI for code building, trust, distrust, familiarity, perceived ease of use, perceived behavioral control, deterrence, complexity.
This investigation paves the way for a comprehensive understanding of how AI is perceived by those who directly manage the introduction of these tools into traditional software development workflows, revealing a road map for future endeavors for the software development community.
Xin Zhao, Brian Vu, Sitesh Pattanaik· AIware· 0 citations
We study the impact of generative artificial intelligence (GAI) tools on product‐level innovation outcomes in the context of software products. Specifically, we illustrate how GAI can alter the direction of innovation by shifting the activities of developers away from generational innovation and toward original innovation, which may be new to the market but not necessarily more novel than previous innovations. We argue that this shift is driven by GAI's ability to facilitate tasks in both the ideation and implementation of software products, which enables some developers to create software products that were previously beyond their reach, while allowing others to reallocate their effort with respect to different activities. Our analyses in the context of browser add‐ons provide empirical evidence for these arguments. We discuss implications for innovation research and the generalizability of our findings to other domains.
Generative AI tools are changing how organizations innovate. Studying the Firefox browser add‐on ecosystem, we find that after ChatGPT's release, the number of new add‐ons increased by 34%, driven by both existing and first‐time developers. However, developers simultaneously reduced improvements to their existing add‐ons by 20%, redirecting effort toward new projects. While new add‐ons drew on a broader set of knowledge domains, they were not substantially different from what already existed. This suggests that generative AI helps developers efficiently combine existing knowledge rather than produce truly novel ideas. For platform managers and business leaders, these findings highlight that although Generative AI can democratize product creation and accelerate new product launches, it may require new strategies to maintain product quality and encourage genuine novelty.
E. Yilmaz, Tim Meyer· Strategic Management Journal· 0 citations
Generative artificial intelligence (GenAI) has moved rapidly from an experimental technology to a tool actively reshaping how managers gather information, evaluate options, and reach decisions across strategic, operational, and customer-facing business functions. This paper reviews the theoretical and empirical literature on GenAI's role in business decision-making, situating recent large language model (LLM)-based applications within the older organizational decision-making and bounded-rationality literature that predates generative AI by decades. The review synthesizes controlled productivity experiments, large-scale economic-potential estimates, organizational decision-structure theory, and the emerging literature on human-AI complementarity and its limits, including evidence that AI assistance can degrade performance when applied outside a model's effective capability frontier. Particular attention is given to the distinction between GenAI's demonstrated value in well-structured, information-synthesis-heavy decision tasks and its more contested role in tasks requiring novel judgment or accountability. Comparative tables summarize reported productivity effects, business function applications, and organizational barriers to adoption across the reviewed literature. The paper concludes that GenAI's current business value is concentrated in augmenting, rather than automating, decision-making, with the strongest evidence for productivity gains among relatively lower-skilled or lower-performing decision-makers, and identifies the mapping of AI capability boundaries within specific decision domains as the central future research prospect. Generative artificial intelligence (GenAI) has moved rapidly from an experimental technology to a tool actively reshaping how managers gather information, evaluate options, and reach decisions across strategic, operational, and customer-facing business functions. This paper reviews the theoretical and empirical literature on GenAI's role in business decision-making, situating recent large language model (LLM)-based applications within the older organizational decision-making and bounded-rationality literature that predates generative AI by decades. The review synthesizes controlled productivity experiments, large-scale economic-potential estimates, organizational decision-structure theory, and the emerging literature on human-AI complementarity and its limits, including evidence that AI assistance can degrade performance when applied outside a model's effective capability frontier. Particular attention is given to the distinction between GenAI's demonstrated value in well-structured, information-synthesis-heavy decision tasks and its more contested role in tasks requiring novel judgment or accountability. Comparative tables summarize reported productivity effects, business function applications, and organizational barriers to adoption across the reviewed literature. The paper concludes that GenAI's current business value is concentrated in augmenting, rather than automating, decision-making, with the strongest evidence for productivity gains among relatively lower-skilled or lower-performing decision-makers, and identifies the mapping of AI capability boundaries within specific decision domains as the central future research prospect.
Rajidi Rammohan Reddy, Vinodray Thumar, Amar Jyoti Borah et al.· International journal of com...· 0 citations
Artificial intelligence (AI) has moved from a peripheral instrument of computational research to an infrastructure that touches nearly every stage of scholarly work. This paper offers an integrative, stage-based analysis of how contemporary AI systems—particularly large language models (LLMs), retrieval-augmented systems, and tool-using agents—are reshaping the research lifecycle, from the earliest formulation of a research question through data analysis, manuscript preparation, peer review, and post-publication dissemination. We propose an eight-stage framework and, for each stage, characterise the dominant AI capabilities in use, the empirical evidence for their benefits, the observed failure modes, and the state of institutional governance. Three cross-cutting findings emerge. First, capability is unevenly distributed across the lifecycle: AI performance is strongest where outputs are cheaply verifiable (code, language, retrieval, structured extraction) and weakest where judgement, tacit knowledge, and problem selection dominate. Second, the principal risks are epistemic rather than merely technical—citation fabrication, homogenisation of research questions, illusions of understanding, and the displacement of accountability—and these risks compound when AI is used at multiple consecutive stages without independent verification. Third, governance has converged rapidly on a narrow consensus (no machine authorship, mandatory disclosure, confidentiality in peer review) while leaving the harder questions of methodological validity and evaluative use largely unresolved. We conclude with a five-level maturity model for responsible institutional adoption and an agenda for research on verification, provenance, and the measurement of AI-attributable scientific value
Dr. Gagandeep Singh, Ravi Ranjan, Anirudh Gupta, Harinakshi Aravind Shetty· International Journal of Adv...· 0 citations
: The rapid advancement of Generative Artificial Intelligence (GAI) is profoundly reshaping the underlying logic and operational modes of enterprise management. Unlike traditional artificial intelligence, GAI not only analyzes and comprehends data but also possesses capabilities such as multimodal content generation, natural language interaction, and adaptive evolution, opening up entirely new possibilities for management innovation. This paper systematically reviews the technological evolution and core capability characteristics of GAI and reveals the micro‑mechanisms by which GAI enables management innovation across three dimensions: knowledge recombination and abductive reasoning, simulation and counterfactual reasoning, and dynamic resource orchestration. Based on a systematic literature search of Web of Science, Scopus, and CNKI (January 2020 – June 2026), 86 relevant sources were identified and synthesized. On this basis, the paper conducts an in‑depth analysis of GAI’s innovative applications in four key scenarios— new product design, customer relationship management, precision marketing, and supply chain management — and incorporates typical domestic and international cases as illustrative examples. The distinctive contribution of this review lies in linking the three mechanisms to dynamic capabilities theory, specifying boundary conditions, and integrating enabling logic with governance frameworks. The findings indicate that while GAI drives management innovation, it also faces multiple challenges, including risks to content quality, technological application, and development. Limitations include reliance on secondary sources and the rapidly evolving technological landscape. Finally, this paper proposes countermeasures from three levels — technical governance, organizational transformation, and institutional safeguards — aiming to provide theoretical references and practical guidance for enterprises to effectively harness GAI and achieve management innovation in the process of intelligent transformation.
Zixuan Li· Business Management Perspect...· 0 citations