Regulating Algorithmic Management of Work and Labor Markets: The Interplay of Public Laws, Private Governance and Direct Negotiations

Processes of recruiting, selecting, hiring, supervising, evaluating, promoting, and dismissing workers and employees are increasingly digitized, automated, and algorithmically mediated, in ways that give rise to concerns about equity, fairness, and the potential for reproducing or even exacerbating inequalities and asymmetries of power. Two major concerns are (1) the opaqueness and difficult explainability of these algorithms and their susceptibility to reproducing the gender, racial, ethnic, and other biases of the vast troves of training data which feed the machine learning that produces these algorithms, and (2) the fact that job applicants and job holders may be subject to arbitrary and unfair decisions about their pay, promotions, and pace and conditions of work dictated by algorithms based on data gathered knowingly or unknowingly from them and about them and about their peers on which they are judged and evaluation. What is the nature of the interplay of the distinct governance mechanisms of formal-legal state regulation, private arrangements, and negotiated bilateral mechanism in shaping governance of AI-based labor markets, labor management, and labor processes? What lessons can be learned from over three decades of experience with such interplay around issues of labor standards in a pre-AI context? How can classic problems of “who will audit the auditors” and misaligned incentives facing for-profit audit firms be addressed? What role can established or novel forms of representation, negotiation, and dialogue play as worker advocates seek transparency and fairness in how algorithms affecting workers are devised and deployed?