Skip to main content

Research Repository

Advanced Search

"What did you say?": Extracting unintentional secrets from predictive text learning systems

Wilkinson, Gwyn; Legg, Phil

"What did you say?": Extracting unintentional secrets from predictive text learning systems Thumbnail


Authors

Gwyn Wilkinson



Abstract

As a primary form of communication, text is used widely in applications including e-mail conversations, mobile text messaging, chatrooms, and forum discussions. Modern systems include facilities such as predictive text, recently implemented using deep learning algorithms, to estimate the next word to be written based on previous historical entries. However, we often enter sensitive information such as passwords using the same input devices-namely, smartphone soft keyboards. In this paper, we explore the problem of deep learning models which memorise sensitive training data, and how secrets can be extracted from predictive text models. We propose a general black-box attack algorithm to accomplish this for all kinds of memorised sequences, discuss mitigations and countermeasures, and explore how this attack vector could be deployed on an Android or iOS mobile device platforms as part of target reconnaissance.

Citation

Wilkinson, G., & Legg, P. (2020, June). "What did you say?": Extracting unintentional secrets from predictive text learning systems. Paper presented at IEEE International Conference on Cyber Security and Protection of Digital Services (Cyber Science 2020)

Presentation Conference Type Conference Paper (unpublished)
Conference Name IEEE International Conference on Cyber Security and Protection of Digital Services (Cyber Science 2020)
Start Date Jun 15, 2020
End Date Jun 19, 2020
Deposit Date May 17, 2020
Publicly Available Date Mar 29, 2024
Public URL https://uwe-repository.worktribe.com/output/5981540
Publisher URL https://www.c-mric.com/csa2020

Files




You might also like



Downloadable Citations