<![CDATA[The rapid expansion of the digital gig economy, driven by transparent algorithmic regimes, has bequeathed a new frontier of re-formation of labor that warrants serious scrutiny. This study investigates how algorithmic bias shapes the dynamic skill formation of freelance workers an overlooked mechanism that influences employability and performance. Employing an explanatory sequential mixed-method design, quantitative questionnaires were administered to 342 gig workers on Upwork, Fiverr, and Gojek platforms and complemented with in-depth interviews of 25 participants. The results show a high positive correlation between perceived algorithmic bias and the dynamics of the demand for skills (β = 0.48, p < .001) which suggests that the higher the perceived bias, the greater the extent of required skillset changes among the workers. The changes are negatively correlated with perceived employability and performance (β = –0.31, p < .001). Qualitative data reveal three interdependent experiences: negotiating the "black box" of algorithmic control, the de-professionalization of vocational skills to secure "algorithmic mastery," and the emergence of adaptive, frequently collective, coping strategies. Synthesizing knowledge from Management Information Systems, Human Capital Theory, and the Technology Acceptance Model, this study constructs an evidence-based model for how algorithmic systems reorganize human capital. The. conclusion emphasizes the importance of transparent algorithmic design and participatory ability-building policy to yield fairness and sustainability in the digital platform of labor.]]>
