Score: 0

Beyond Automation: Rethinking Work, Creativity, and Governance in the Age of Generative AI

Published: December 9, 2025 | arXiv ID: 2512.11893v1

By: Haocheng Lin

Potential Business Impact:

AI changes jobs, creativity, and how we live.

Business Areas:
Artificial Intelligence Artificial Intelligence, Data and Analytics, Science and Engineering, Software

The accelerating advancement of generative artificial intelligence (AI) systems is reshaping the nature, distribution and meaning of work, creativity, and economic security. This paper investigates four inter-related phenomena in the current AI era: (1) the evolving landscape of employment and the future of work; (2) the diverse patterns of AI adoption across socio-demographic groups, sectors, and geographies; (3) whether universal basic income (UBI) should become a compulsory policy response to the AI revolution; and (4) the implications of AI content policies and model behaviours for human creativity, wellbeing, and everyday decision-making. Furthermore, the paper tests the hypothesis that newer model generations may perform worse than their predecessors, and examines how users' interactions with AI systems may produce echo chambers through sycophantic model alignment. Using a mixed methodology that integrates labour market task-exposure modelling, sectoral diffusion mapping, policy-framework analysis, and qualitative discourse critique, this study develops a comprehensive framework for understanding the societal consequences of AI systems beyond productivity gains. It argues that to foster an inclusive, meaningful, and creative environment, policymakers must treat UBI as one dimension within a broader ecosystem of governance, skills development, creativity preservation, and model design. The paper concludes by outlining future research directions, including systematic evaluation of AI's creative performance across model generations, construction of a taxonomy of AI-usage distribution and equity, and formulation of governance criteria to balance content restrictions with creative freedom.

Page Count
47 pages

Category
Computer Science:
Computers and Society