William Seymour, King's College London
The conundrum of how voice assistants should broker consent to share data with third party software has proven to be a tough problem without a clear solution. Contemporary assistants often require users to switch to other devices in order to navigate permissions menus for their otherwise hands-free voice assistant, as with smartphone apps. More in line with modern smartphones, Alexa now offers "voice-forward consent", allowing users to grant skills access to personal data mid-conversation using speech.
While more usable and convenient than opening a companion app, asking for consent 'on the fly' can undermine several concepts core to the consent process. The intangible nature of voice interfaces also blurs the boundary between parts of an interaction that are controlled by third-party developers from the underlying platforms. This talk draws on the GDPR and work on consent in HCI and Ubicomp to outline and address the problems with brokering consent verbally.
William Seymour, King's College London
William Seymour is a researcher at King's College London, exploring ways of explaining the privacy and security behaviours of AI assistants as part of the EPSRC-funded Secure AI Assistants project. Before this he was part of a project funded by the UK data protection regulator on the future of data protection in smart homes. William recieved his PhD in Cyber Security from the University of Oxford for research exploring ethical issues with IoT devices.
author = {William Seymour},
title = {Consent on the Fly: Developing Ethical Verbal Consent for Voice Assistants},
year = {2022},
address = {Santa Clara, CA},
publisher = {USENIX Association},
month = jun
}
