On the Trustworthiness Landscape of State-of-the-art Generative Models: A Survey and Outlook

TL;DR

This survey examines the trustworthiness issues of state-of-the-art generative models like diffusion and large language models, analyzing risks across privacy, security, fairness, and responsibility, and offering future research directions.

Contribution

It provides a comprehensive map of trustworthiness challenges in large generative models and suggests practical recommendations and research directions for enhancing their security and societal benefit.

Findings

Identified key trustworthiness threats in privacy, security, fairness, and responsibility.

Developed an extensive map outlining the trustworthiness landscape.

Proposed practical recommendations and future research directions.

Abstract

Diffusion models and large language models have emerged as leading-edge generative models, revolutionizing various aspects of human life. However, the practical implementations of these models have also exposed inherent risks, bringing to the forefront their evil sides and sparking concerns regarding their trustworthiness. Despite the wealth of literature on this subject, a comprehensive survey specifically delving into the intersection of large-scale generative models and their trustworthiness remains largely absent. To bridge this gap, this paper investigates both the long-standing and emerging threats associated with these models across four fundamental dimensions: 1) privacy, 2) security, 3) fairness, and 4) responsibility. Based on the investigation results, we develop an extensive map outlining the trustworthiness of large generative models. After that, we provide practical recommendations and potential research directions for future secure applications equipped with large generative models, ultimately promoting the trustworthiness of the models and benefiting the society as a whole.