Develop a set of specific recommendations for a particular policy-making body anywhere in the world about how they should regulate a particular application area of machine learning (e.g., hiring, bail setting, sentencing, drone targeting, autonomous vehicles, use of facial recognition by law enforcement, etc.)

Develop a set of specific recommendations for a particular policy-making body anywhere in the world about how they should regulate a particular application area of machine learning (e.g., hiring, bail setting, sentencing, drone targeting, autonomous vehicles, use of facial recognition by law enforcement, etc.) Develop a set of thoughtful, nuanced recommendations that encompass (1) any circumstances under which you think the use of automated decision-making should be prohibited in this application area, (2) what requirements, if any, should be imposed on the data sets used to develop and train the models, (3) who should have oversight over the final decisions and what that oversight entails, (4) who should be responsible for errors, security breaches, or malfunctions of the automated decision making tools and what the potential consequences for those entities are, (5) what specific transparency or audit requirements, if any, the developers of the automated decision-making tools will be subject to, and (6) how the decision-making tools will be assessed for bias or discrimination, if at all, and who will be responsible for performing those assessments.