2020/02/09 by Filipo Sharevski, Sharevski, Filipo, Peter Jachim +9 · 1 voice
Computer Science · Social Sciences · #Hate Speech and Cyberbullying Detection #Misinformation and Its Impacts #Privacy, Security, and Data Protection
paper · pdf · doi:10.1016/j.ijhcs.2021.102604
openalex publication_date 2021/01/28 · crossref created 2021/01/28 · crossref issued 2021/05/01 · crossref published 2021/05/01 · crossref published-print 2021/05/01 · openalex created_date 2025/10/10 · crossref deposited 2025/10/21 · crossref indexed 2025/11/11 · openalex updated_date 2026/07/28
This paper reports the findings of a study where users (N=220) interacted\nwith Malexa, Alexa's malicious twin. Malexa is an intelligent voice assistant\nwith a simple and seemingly harmless third-party skill that delivers news\nbriefings to users. The twist, however, is that Malexa covertly rewords these\nbriefings to intentionally introduce misperception about the reported events.\nThis covert rewording is referred to as a Malware-Induced Misperception (MIM)\nattack. It differs from squatting or invocation hijacking attacks in that it is\nfocused on manipulating the "content" delivered through a third-party skill\ninstead of the skill's "invocation logic." Malexa, in the study, reworded\nregulatory briefings to make a government response sound more accidental or\nlenient than the original news delivered by Alexa. The results show that users\nwho interacted with Malexa perceived that the government was less friendly to\nworking people and more in favor of big businesses. The results also show that\nMalexa is capable of inducing misperceptions regardless of the user's gender,\npolitical ideology or frequency of interaction with intelligent voice\nassistants. We discuss the implications in the context of using Malexa as a\ncovert "influencer" in people's living or working environments.\n