An unethical optimization principle
If an artificial intelligence aims to maximize risk-adjusted return, then under mild conditions it is disproportionately likely to pick an unethical strategy unless the objective function allows sufficiently for this risk. Even if the proportion η of available unethical strategies is small, the prob...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
The Royal Society
2020-07-01
|
Series: | Royal Society Open Science |
Subjects: | |
Online Access: | https://royalsocietypublishing.org/doi/pdf/10.1098/rsos.200462 |