This paper investigates how Artificial Intelligence (AI) goes beyond traditional technological roles to drive transformation across industries, organisations, and society. By analysing 2188 academic papers from the arXiv platform (from 2018 until 2025), the study highlights the unexpectedly widespread adoption of AI and its increasing specialisation in tasks and organisational functions. The findings reveal AI's potential to foster innovation, enhance productivity, and address complex challenges. While the focus on academic sources and the exclusion of regional contexts present certain limitations, this research provides critical insights into AI's evolving role and its implications for businesses, employees, and the conduct of government. Purpose/aims of the paper This study explores how AI has evolved from technical applications to become a key driver of innovation and knowledge creation. It investigates AI's penetration across industries, organisational areas, and tasks, focusing on its ability to solve complex challenges, reshape business workflows, and enable strategic decision-making in diverse contexts. Research methodology The research employed secondary data analysis of 2188 full-text academic papers sourced from the arXiv platform. Structured keyword extraction, conducted using Large Language Models, identified patterns and relationships between industries, organisational areas, and tasks. Additionally, trends in AI adoption, specialisation, and integration were examined to uncover its transformative impact on organisations and knowledge systems. Findings/conclusions The findings reveal AI's significant penetration into industries and its increasing specialisation in organisational areas and tasks. Beyond automation, AI fosters task-specific solutions and innovation, addressing organisational challenges and improving productivity. These findings underscore AI's transformative potential to redefine business practices, enhance collaboration, and drive societal progress by pushing the boundaries of innovation and knowledge creation. Discussion AI's influence has expanded far beyond technical functions, establishing itself as a strategic tool for innovation, cross-disciplinary problem-solving, and knowledge generation. Its integration into diverse sectors reflects its capacity to foster collaboration and address societal and organisational challenges. Adapting to AI's rapid evolution is crucial for businesses, employees, and governments navigating this dynamic landscape. Research limitations This study relied on 2188 academic papers from the arXiv platform, centring its findings on academic contexts. Moreover, regional and cultural differences in AI adoption were not addressed. Future research should incorporate broader datasets and interdisciplinary approaches to provide a more comprehensive understanding of AI's global impact and socio-economic implications. Practical implications/applications to practice Organisations can leverage AI to drive productivity, optimise processes, and create innovative solutions across industries. Strategic AI adoption requires investment in workforce reskilling and the ethical implementation of AI systems. Decision-makers must prioritise long-term planning to achieve AI's full potential while ensuring it supports both organisational competitiveness and sustainable development. Social implications/impact on society and/or policy AI's rapid expansion highlights its potential to address societal challenges, from reducing inequities to fostering global development. However, pressing issues such as ethical use, algorithmic fairness, and data privacy demand urgent attention. Governments must establish equitable frameworks to ensure that AI supports inclusive progress and mitigates potential negative consequences, fostering sustainable societal progress. Originality/what is new about your research? This study uniquely combines large-scale empirical evidence with theoretical perspectives to analyse AI adoption trends. It highlights AI's penetration into industries, organisational areas, and tasks, emphasising its progression beyond traditional applications to drive task-specific specialisation, reshape innovation processes, and expand the boundaries of knowledge creation.
