Truthful AI: Developing and governing AI that does not lie

Authors

Owen Cotton-Barratt, Max Daniel, William MacAskill

Abstract

This paper discusses the importance of truthfulness in AI systems and proposes approaches to ensure AI systems provide accurate information.

Publication Details

Published:

December 5th, 2020

Venue:

arXiv

Added to AI Safety Papers:

March 16th, 2025

Metadata

Tags:

TruthfulnessAI SafetyGovernance

Original Paper:

Link