Working Paper: NBER ID: w29267
Authors: Jens Ludwig; Sendhil Mullainathan
Abstract: Algorithms (in some form) are already widely used in the criminal justice system. We draw lessons from this experience for what is to come for the rest of society as machine learning diffuses. We find economists and other social scientists have a key role to play in shaping the impact of algorithms, in part through improving the tools used to build them.
Keywords: Algorithms; Criminal Justice; Bias; Decision-making
JEL Codes: C01; C54; C55; D8; H0; K0
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| poorly constructed algorithms (C69) | exacerbate bias (D91) |
| well-built algorithms (C63) | reduce bias (C46) |
| human decisions in algorithm construction (C44) | observed failures or successes (C90) |
| poorly constructed algorithms (C69) | failures in the criminal justice system (K40) |
| biased data (C83) | exacerbate racial bias (J15) |
| flawed design choices (L15) | exacerbate racial bias (J15) |
| introduction of algorithms (Y20) | new problems (C62) |
| fragility of algorithms (C69) | mispredictions and inconsistent outcomes (D80) |
| design and implementation of algorithms (C90) | effectiveness and fairness of algorithms (D63) |