A Survey on Responsible Generative AI: What to Generate and What Not

TL;DR

This survey reviews the current state of responsible generative AI, focusing on safety, ethics, and practical considerations for ensuring models generate truthful, non-toxic, and non-leaking content across various domains.

Contribution

It provides a comprehensive overview of responsible requirements for both textual and visual generative models, highlighting recent advancements and challenges in safety and ethics.

Findings

Identifies key responsible generation requirements: truthfulness, toxicity avoidance, harm refusal, data privacy, and identifiability.

Discusses the importance of responsible GenAI in healthcare, education, finance, and AGI.

Provides insights into safety challenges and future directions for responsible generative AI.

Abstract

In recent years, generative AI (GenAI), like large language models and text-to-image models, has received significant attention across various domains. However, ensuring the responsible generation of content by these models is crucial for their real-world applicability. This raises an interesting question: What should responsible GenAI generate, and what should it not? To answer the question, this paper investigates the practical responsible requirements of both textual and visual generative models, outlining five key considerations: generating truthful content, avoiding toxic content, refusing harmful instruction, leaking no training data-related content, and ensuring generated content identifiable. Specifically, we review recent advancements and challenges in addressing these requirements. Besides, we discuss and emphasize the importance of responsible GenAI across healthcare, education, finance, and artificial general intelligence domains. Through a unified perspective on both textual and visual generative models, this paper aims to provide insights into practical safety-related issues and further benefit the community in building responsible GenAI.