On the Ethics of Building AI in a Responsible Manner
Machine Learning
2020-04-10 v1
Abstract
The AI-alignment problem arises when there is a discrepancy between the goals that a human designer specifies to an AI learner and a potential catastrophic outcome that does not reflect what the human designer really wants. We argue that a formalism of AI alignment that does not distinguish between strategic and agnostic misalignments is not useful, as it deems all technology as un-safe. We propose a definition of a strategic-AI-alignment and prove that most machine learning algorithms that are being used in practice today do not suffer from the strategic-AI-alignment problem. However, without being careful, today's technology might lead to strategic misalignment.
Keywords
Cite
@article{arxiv.2004.04644,
title = {On the Ethics of Building AI in a Responsible Manner},
author = {Shai Shalev-Shwartz and Shaked Shammah and Amnon Shashua},
journal= {arXiv preprint arXiv:2004.04644},
year = {2020}
}