Algorithmic Bias And Accountability In Predictive Policing: A Legal And Ethical Appraisal
Himadri Mishra, Amity University, Amity Law School, Lucknow
ABSTRACT
As a data-oriented law enforcement method, predictive policing combines statistical methods, machine learning models, and massive datasets to forecast crime trends and determine the actions of law enforcement. Although it is marketed as a strategy to boost efficiency and objectivity, predictive policing systems have received a lot of legal and ethical controversy because they may replicate and magnify the structural biases that are inherent in previous crime data. In this research paper, the antecedents and symptoms of algorithmic bias in predictive policing systems like COMPAS, PredPol, and LASER are critically analyzed and the impacts that the systems have on decision-making and operational policing practices are discussed.
The paper provides an assessment of the regulatory responses that have been adopted in different jurisdictions such as the United States, the United Kingdom, the European Union, Japan, and India through a doctrinal and comparative approach methodology backed by empirical evidence that is in the form of audit reports and experimental studies. The major court cases such as State v. Loomis and Bridges v. Chief Constable of South Wales Police, show that there is an urgent necessity to apply transparency, accountability, and rights-based supervision over automated decision- making within the criminal justice system. The paper finds that predictive policing algorithmic bias is not only a technical flaw, but a structural issue that requires a sound legal amendment, improved governance structures, and technologically responsible design.
Keywords: Predictive policing, algorithm bias, accountability, artificial intelligence, machine learning, criminal justice, due process, discrimination, data protection, legal regulation.
