Immunization Among Voice Assistant Users

“Hey Alexa, What do You Know About the
COVID-19 Vaccine?” – (Mis)perceptions of Mass
Immunization Among Voice Assistant Users
No Author Given
No Institute Given
Abstract. In this paper, we analyzed the perceived accuracy of COVID19 vaccine information spoken back by Amazon Alexa. Unlike social media, Amazon Alexa doesn’t apply soft moderation to unverified content,
allowing for use of third-party skills to arbitrarily phrase information
about the COVID-19 vaccine. The results from a 210-participant study
with Amazon Alexa suggest that a malicious third-party skill is successful in reducing the perceived accuracy of information as to who gets the
vaccine first, vaccine testing, and the side effects of the vaccine. We also
found that the vaccine-hesitant participants are drawn to pessimistically
rephrased Alexa responses focused on the downsides of the mass immunization. We discuss solutions for soft moderation against misperceptioninducing or COVID-19 misinformation malicious skills.
Keywords: Alexa · Malicious Skills · COVID-19 Vaccines
1 Introduction
Vaccine skepticism, with the extreme of anti-vaccination or anti-vax sentiment,
roots itself in fear and uncertainty about the safety, efficacy, and side effects
of vaccines. Safety concerns, fueled by unsupported claims or exaggerated facts
of side effects [6], can increase the public’s receptivity to conspiracy theories
or alternative treatments to avoid perceived risk [28]. In the absence of definitive authoritative information, the spread of such narratives can detrimentally
impact public health, particularly if recent events have eroded public trust in
vaccines. Unfortunately, though anti-vax communities on social media platforms
like Facebook and Twitter are small in both number and size, their engagement
with undecided users on those platforms is high, particularly compared to that
of pro-vaccination groups [28]. Even when users search for vaccine information
via search engines, they may still fall into the pitfalls of anti-vax content, “selfreferencing and mutually reinforcing links that can fool users into believing”
these ideas are widely held and plausible [30]. As users prioritize online sources
over the advice of health professionals in their medical decision-making [30], the
success and persistence of vaccine scepticism and anti-vax narratives grows more
troubling, threatening the global combat against the COVID-19 pandemic.
Most recently, the COVID-19 pandemic forced the authorities to take unprecedented steps to develop, test, and disseminate a vaccine in a time-frame an
2 No Author Given
order of magnitude faster then the normal course for developing and approving
vaccines for viruses [54]. The extreme emergency for inoculation, exacerbated
by the political (mis)use of the pandemic in the US [18], catalyzed the spread
of alternative narratives about the COVID-19 vaccine focused on the vaccine
safety, gaps in testing, and serious side effects [5]. The potency of these narratives and the sheer volume of misinformation prompted the World Health
Organization (WHO) and the Centers for Disease Control (CDC) to maintain
a “myth-busters” section on their websites about the COVID-19 vaccine and
virus[57] [16]. It also forced social media platforms like Twitter to apply “soft
moderation,” e.g. label tweets with misleading or harmful information that could
“incite people to action and cause widespread panic, health anxiety, and fear that
could lead to social unrest or large-scale disorder” [44], [51].
Consequently, most of the academic attention about the alternative narratives of the COVID-19 vaccine has focused on the dissemination of misinformation on mainstream social media [5], [42], [26] (some work has been done on
alternative social media too, e.g. Parler [1], [38] where COVID-19 misinformation, unfortunately, is rampant). Mainstream social media platforms allow for
visual discernment of the information, formation of so-called “influencer” accounts, and direct communication of the engagement with the content metrics
such as number of replies, re-tweets, likes, and shares. While all of these factors
certainly affect the receptivity of any COVID-19 vaccine information posted on
these platforms, or any website for further visual inspection, little attention is
devoted to exploring how people respond to both COVID-19 vaccine narratives
when these are delivered through a voice assistant like Amazon Alexa.
Unlike social media, Alexa is the sole authority or “influencer” that delivers
information when prompted without disclosing the source of the information
of any engagement metrics (if any). Studies in the past had found that users
usually trust Alexa and worry only about Alexa intruding into their privacy,
but not about the validity of the information delivered by Alexa [31]. Akin
to posting unverified claims on mainstream social media, studies have shown
that bad actors can develop malicious third-party applications, called “skills”
for Alexa that can silently rephrase information from any source to mislead a
user and induce misperception about a polarizing topic such as vaccination, free
speech, or government actions [46],[49],[48]. This motivated us to explore how
users will respond when such a third-party skill is used to deliver rephrased
information about the COVID-19 vaccine.