Typology of Risks of Generative Text-to-Image Models
Typology of Risks of Generative Text-to-Image Models
复制标题
DOI：
10.1145/3600211.3604722
复制
发表时间：
2023-07
期刊：
影响因子：
--
通讯作者：
Charlotte M. Bird;Eddie L. Ungless;Atoosa Kasirzadeh
中科院分区：
文献类型：
--
作者：
Charlotte M. Bird;Eddie L. Ungless;Atoosa Kasirzadeh
本文通过全面的文献综述，研究了与现代文本到图像生成模型（如DALL-E和Midjourney）相关的直接风险和危害。虽然这些模型提供了前所未有的生成图像的能力，但它们的开发和使用引入了需要仔细考虑的新型风险。我们的审查显示，尽管一些已经得到解决，但在理解和治疗这些风险方面存在重大知识差距。我们提供了六个主要利益相关者群体的风险分类，包括未探索的问题，并建议未来的研究方向。我们确定了22种不同的风险类型，涵盖了从数据偏见到恶意使用的问题。这里介绍的调查是为了加强正在进行的负责任的模型开发和部署的话语。通过强调以前被忽视的风险和差距，它旨在塑造后续的研究和治理计划，引导他们走向负责任，安全和道德意识的文本到图像模型的演变。
This paper investigates the direct risks and harms associated with modern text-to-image generative models, such as DALL-E and Midjourney, through a comprehensive literature review. While these models offer unprecedented capabilities for generating images, their development and use introduce new types of risk that require careful consideration. Our review reveals significant knowledge gaps concerning the understanding and treatment of these risks despite some already being addressed. We offer a taxonomy of risks across six key stakeholder groups, inclusive of unexplored issues, and suggest future research directions. We identify 22 distinct risk types, spanning issues from data bias to malicious use. The investigation presented here is intended to enhance the ongoing discourse on responsible model development and deployment. By highlighting previously overlooked risks and gaps, it aims to shape subsequent research and governance initiatives, guiding them toward the responsible, secure, and ethically conscious evolution of text-to-image models.