Mitigating Covertly Unsafe Text within Natural Language Systems

Alex Mei, Anisha Kabir, Sharon Levy, Melanie Subbiah, Emily Allaway, John Judge, Desmond Patton, Bruce Bimber, Kathleen McKeown, William Yang Wang


Abstract
An increasingly prevalent problem for intelligent technologies is text safety, as uncontrolled systems may generate recommendations to their users that lead to injury or life-threatening consequences. However, the degree of explicitness of a generated statement that can cause physical harm varies. In this paper, we distinguish types of text that can lead to physical harm and establish one particularly underexplored category: covertly unsafe text. Then, we further break down this category with respect to the system’s information and discuss solutions to mitigate the generation of text in each of these subcategories. Ultimately, our work defines the problem of covertly unsafe language that causes physical harm and argues that this subtle yet dangerous issue needs to be prioritized by stakeholders and regulators. We highlight mitigation strategies to inspire future researchers to tackle this challenging problem and help improve safety within smart systems.
Anthology ID:
2022.findings-emnlp.211
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2022
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2914–2926
Language:
URL:
https://aclanthology.org/2022.findings-emnlp.211
DOI:
10.18653/v1/2022.findings-emnlp.211
Bibkey:
Cite (ACL):
Alex Mei, Anisha Kabir, Sharon Levy, Melanie Subbiah, Emily Allaway, John Judge, Desmond Patton, Bruce Bimber, Kathleen McKeown, and William Yang Wang. 2022. Mitigating Covertly Unsafe Text within Natural Language Systems. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 2914–2926, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
Mitigating Covertly Unsafe Text within Natural Language Systems (Mei et al., Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-emnlp.211.pdf
Video:
 https://aclanthology.org/2022.findings-emnlp.211.mp4