Generative AI and Social Transformation: A Socio-Technical Framework for Trust, Identity, and Accountability in Human–AI Collaboration
Main article
Abstract
Generative artificial intelligence (GenAI) has rapidly moved from a research curiosity to a general-purpose technology that mediates an expanding share of professional knowledge work, public service delivery, and everyday social interaction. The acceleration creates a coordination problem that purely technical evaluations cannot address: workers, organizations, and institutions must decide when to trust model outputs, how to renegotiate professional identities that overlap with machine competencies, and where to locate responsibility when generative systems contribute to consequential decisions. This article develops a socio-technical framework that links three interdependent dimensions of human–AI collaboration—trust calibration, identity work, and distributed accountability—and shows how the framework operates across four high-stakes domains: healthcare, education, knowledge work, and public governance. Drawing on a systematic synthesis of 247 peer-reviewed studies published between 2015 and 2025, the article reconstructs the empirical patterns that have emerged after the public release of large language models, identifies the regulatory and organizational arrangements that condition the diffusion of GenAI, and proposes an adaptive governance layer that connects the three dimensions to broader institutional structures. Three integrative propositions emerge from the analysis. First, trust is best understood as a calibrated relation that organizations can engineer through verification routines rather than as a static disposition. Second, identity work is the central socio-cognitive labor that determines whether GenAI augments or displaces human expertise. Third, accountability for generative output should be distributed across designers, operators, institutions, and affected publics in proportion to their causal and epistemic contributions. The article concludes with a research agenda that highlights longitudinal designs, cross-cultural comparison, and multi-level integration between micro practices and macro institutional conditions.
