Any true capacity for ethical reasoning those systems exhibit would be sort of incidental — they’re just predictors of how human users respond to questions, and they’ll use any approach they stumble on that has good predictive value.
– Kelsey Piper
Vox reports on an experimental AI model called Delphi that tries to predict human responses to moral situations. Development work on Delphi is being done at the Allen Institute for AI in Seattle.
How well can an AI mimic human ethics? | VOX | October 27, 2021 | by Kelsey Piper