Machine learning algorithms and artificial intelligence influence many aspects of life today and have gained an aura of objectivity and infallibility. The use of these tools introduces a new level of risk and complexity in policy. This report illustrates some of the shortcomings of algorithmic decisionmaking, identifies key themes around the problem of algorithmic errors and bias, and examines some approaches for combating these problems.
RAND has contributed more to thinking about how to deal with the longer-range future than any other organization. The methodological work at RAND started in the late 1950s and 1960s with the developments of the Delphi method and scenario analysis. This work continued sporadically through the 1970s and 1980s—and since the 1990s—computational technology has enabled new futures methodologies such as Robust Decision Making and Scenario Discovery.
RAND is again leading the way in putting these methods to use in dealing with the challenges of the future. Below is a list of publications that exemplifies RAND's work on Robust Decision Making methodology and applications.
Testing the Scenario Hypothesis: An Experimental Comparison of Scenarios and Forecasts for Decision Support in a Complex Decision Environment March 16, 2017
Decision support tools are known to influence and facilitate decisionmaking through the thoughtful construction of the decision environment. However, little research has empirically evaluated the effects of using scenarios and forecasts.
Evaluation of the Jinan City Water Ecological Development Implementation Plan and Recommendations for Improvement March 9, 2017
RAND evaluated potential effects of uncertain projections of demand and climate change on the ability of the Jinan Municipal Water Resources Bureau to meet its long-term water resources goals. This document describes RAND's approach and results, including development of a mathematical simulation model of the Jinan water supply system and analysis of the system's performance if new strategies and investments were to be implemented.
Urban Responses to Climate Change: Framework for Decisionmaking and Supporting Indicators December 7, 2016
This report presents a decisionmaking framework and supporting indicators to guide urban responses to climate change based on principles of risk governance, along with considerations of institutional arrangements and their economic and social context. The framework emphasizes identification of indicators of capacities and processes to implement, adapt, and transform policies, institutions, financing, and other actions to affect change.
Improving Decision Support for Infectious Disease Prevention and Control: Aligning Models and Other Tools with Policymakers' Needs July 12, 2016
This report provides an overview of decision-support tools, including models and nonmodeling approaches, to inform both modelers and policymakers. It offers ways to use these tools to help address real-world policy questions related to infectious disease prevention, detection, control, and response and recommends best practices for technical experts and policymakers to collaborate in developing and using the tools.
Contestability Frameworks: An International Horizon Scan March 21, 2016
RAND identified and reviewed international contestability practices on behalf of the Australian Department of Defence (ADoD), which is restructuring its military capability acquisition process. A key component of this restructuring is the establishment of an internal contestability capability to assess ADoD's requirements, acquisition, and budget decisions internally before they are passed to other elements in the government.
Scenarios are widely used for long-term climate and energy analysis.
Uncertainty-Sensitive Heterogeneous Information Fusion: Assessing Threat with Soft, Uncertain, and Conflicting Evidence February 1, 2016
This report presents research on various methods for heterogeneous information fusion — combining data that are qualitative, subjective, fuzzy, ambiguous, contradictory, and even deceptive, in order to form a realistic uncertainty-sensitive assessment of threat. The context is counterterrorism, for both military and civilian applications, but the ideas are more generally applicable in intelligence and law enforcement.