Assessing Security and Privacy Insights for Smart Home Users
Samiah Alghamdi
a
and Steven Furnell
b
Cyber Security Research Group, School of Computer Science, University of Nottingham, Nottingham, U.K.
Keywords: Smart Home, Privacy, Security, Internet of Things, Usability.
Abstract: Recently, the number and range of Internet-connected devices have increased rapidly, especially due to
adoption of the Internet of Things and smart home contexts. As a result, users can find themselves needing to
be concerned with the security and privacy of an increasing range of devices. This paper explores the
challenges that users can face in understanding and using the related features on their devices. The first
element of the work is approached by assessing the user-facing materials (e.g., instruction manuals and online
guidance) for a wide variety of smart home devices to determine the extent to which security and privacy
aspects (and related features) are highlighted and explained. Having established that the situation is
inconsistent, the work proceeds to assess the user experience in practice, by examining how easily a series of
security and privacy-related tasks may be accomplished via three alternative smart speakers. The findings
highlight further inconsistency and suggest that users could face considerable challenges keeping track of
security settings and status of multiple devices across a smart home, and the need for information to be
presented in a more coherent form.
1 INTRODUCTION
Smart homes are based upon Internet-connected
versions of household devices that have traditionally
operated in a standalone manner. Satpathy (2006)
described a smart home as one "which is smart
enough to assist the inhabitants to live independently
and comfortably with the help of technology is
termed as a smart home. In the smart home, all the
mechanical and digital devices are interconnected to
form a network, which can communicate with each
other and the user to create an interactive space".
Smart home devices may share information
gathered by applications and can exist in widely
varying configurations. This, in turn, indicates that
devices can be connected to one or many devices.
While users are offered resulting flexibility and
convenience, the fact that the devices are online, as
well as collecting, storing, and communicating user
data, leads to associated considerations around
security and privacy.
The rapid growth and proliferation of smart home
devices has made it difficult for some users to keep
up with the pace of change. While they use the
a
https://orcid.org/0000-0003-3028-2910
b
https://orcid.org/0000-0003-0984-7542
devices, they may not fully understand and manage
the resulting security and privacy options that may be
associated with them. With this in mind, this research
aims to investigate the user experience and determine
areas in which further support may be needed. The
main contributions of the resulting study are:
to investigate the extent to which security and
privacy issues are made apparent to users of
smart home devices.
to assess the nature of the user experience when
attempting to utilise smart home devices to
perform security- and privacy-related operations.
2 BACKGROUND
Today's common smart home includes smart TVs,
speakers, cameras, music streaming devices, smart
lighting, and smart thermostats. Moreover, many
smart home devices may be linked to a managing
system utilizing a house location network. Various
devices interact with the user's phone applications
and communicate with remotely hosted services
(Mazwa & Mazri, 2018). For instance, many security
592
Alghamdi, S. and Furnell, S.
Assessing Security and Privacy Insights for Smart Home Users.
DOI: 10.5220/0011741800003405
In Proceedings of the 9th International Conference on Information Systems Security and Privacy (ICISSP 2023), pages 592-599
ISBN: 978-989-758-624-8; ISSN: 2184-4356
Copyright
c
2023 by SCITEPRESS Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)
cameras keep video storage on an external server that
users can access anywhere and anytime. Meanwhile,
smart assistants, such as Alexa, benefit from the use
of cloud-based benefits.
Many researchers use the Internet of Things
concept to classify security issues in smart homes into
four categories: application security, device security,
network security, and cloud-based security (Alrawi et
al., 2019). For clarification, smart home devices are
advanced IoT devices that can sense data and record
and transmit it to servers hosted on the cloud,
delivering the services to smart home users. For
instance, when a security camera detects motion, a
video will be saved and transferred to the cloud for
further analysis or storage. Additionally, most
devices provide smartphone apps that permit users to
interact with and configure them.
Information security means protecting data from
unauthorized access. Privacy concerns involve a
user's control over the creation and usage of their data
(Kotz et al., 2009). Consequently, privacy depends on
security, as the data must be genuinely secure before
privacy concerns arise. However, privacy is essential
for smart home applications, as there are many users
influencing the same information, and the connection
between them is required to clearly explain and
clarify the issues of data ownership. The domain of
suggested solutions for IoT privacy covers network,
design, and socio-technical efforts; Nevertheless,
Jacobsson & Davidsson (2015) recommend that it is
essential to understand users to be able to generate
usable privacy tools. This motivates us to assess the
availability of privacy and security information to
users if they search for it.
Previous HCI researchers have investigated the
mental models of users using smart home devices. For
example, Zheng (2017) conducted semi-structured
interviews with the users of smart homes were
performed to comprehend their privacy concerns and
mental models. They discovered that users of smart
homes sacrifice their privacy for their convenience.
However, Emami-Naeini et al., (2019) and Leo
Gorski et al. (2018) investigated smart home security
perceptions, determined characteristics that impact
security decisions, and studied users' concerns before
and after buying devices. Other researchers
investigated access-control procedures for smart
home devices (Colnago et al., 2020; He et al., 2018;
Zeng et al., 2019). For example, Zeng et al. (2019)
created a prototype and assessed the usability of an
access control application. Colnago et al. (2020)
analysed Personalised Privacy assistants in the IoT to
permit users to discover and manage data collection
with nearby smart devices. He et al., (2018) explored
how access control policies in smart home devices
vary based on contextual aspects such as device
abilities and user relationships.
Many user privacy experiments concerning IoT
technologies have been performed in temporary or
laboratory environments. One such experiment was
conducted on five users in one week with a custom
IoT device (Worthy et al., 2016). Moreover,
experiments conducted to understand user concerns
privacy with smartwatches (Udoh & Alkharashi,
2017) and toys connected to the internet (Mcreynolds
et al., 2013) have also explained user attitudes and
identified more functional designs for IoT privacy.
Related work had a few privacy concerns regarding
the data itself in nature; with apparent concerns about
how businesses would manage the data (Rodden et
al., 2013) Participants in this experiment were
principally interested in increasing advertising and
marketing their data for profit.
Based upon the resulting appreciation of the
problem, it is considered that users would benefit from
a more effective exposition of security and privacy
issues, as well as a harmonised means of understanding
the related status of their smart home and associated
devices within it. These issues are consequently
explored in more detail in the following sections.
3 EVALUATING PROVISION OF
INFORMATION
Although services provided by smart home devices
improve our quality of life, they invariably increase
concerns regarding the privacy of personal
information. It is therefore relevant to consider the
extent to which security and privacy issues are
highlighted to users, and the extent to which they
must search to find relevant information. With this in
mind, we evaluated what users could find out about
the devices from online sources and the extent to
which users are presented with this information as
part of the standard guidance. The evaluation
considered six types of smart home devices: TVs,
Speakers, Thermostats, Robotic Vacuum, Smart
Video Doorbells and Displays. We then evaluated an
average of three companies for each type.
Furthermore, we selected companies with large,
medium, and small market shares in smart homes. In
each case we examined whether security/privacy
aspects were mentioned in the manufacturer’s
product web page, the user manual, an online FAQ
page (or similar) or other online sources. The latter
would include the privacy policies of smart home
Assessing Security and Privacy Insights for Smart Home Users
593
devices. Multiple companies take the form of long
privacy policies, which usually include legal jargon
and are difficult for users to read (Fabian et al., 2017),
making it cumbersome, if not inconceivable, for them
to use and control their data effectively.
The sub-sections that follow present a summary
of the findings for each category of smart device.
3.1 Smart Speakers
Smart speakers enable users to control and interact
with the device via voice commands, using a
virtual/voice assistant (Alexa, Google, Siri) to
perform tasks and access online services.
This assessment evaluates Google Nest, Amazon
Echo Dot, and Apple HomePod. Most devices and
their applications have the potential to invade the
user's privacy by collecting personal data, addresses,
voice recordings, and the geolocation of the user's
smartphone. Moreover, most smart speakers share
voice recordings with third parties for purposes such
as marketing and improving functionally (the
HomePod, did not share data with third parties).
Regarding privacy-related information presented
in the user manual page, we found that Google Nest
speakers indicated how to turn off the microphone.
The HomePod presented more information about user
privacy than other smart speakers. It will be difficult
for the users to read and understand privacy policy
pages because they were over long and complicated.
3.2 Smart Thermostat
Smart thermostats are a recent technology that
connects the heating system in a home to the Internet,
allowing the user to change the temperature or turn
off the heating from anywhere using the application.
This review examined the Netatmo smart ther-
mostat and the Amazon smart thermostat. Most smart
thermostat and their applications have the potential to
invade the user's privacy. All smart thermostats collect
personal information. Furthermore, Netatmo can
collect the house location, whereas Amazon collects
voice recordings because the user will require the
speaker to control the device via voice activation.
We found that all devices do not present privacy-
related information on the user manual page. Again,
the privacy policy pages are too long and complicated
for users to read and understand.
3.3 Robotic Vacuum
A robotic vacuum is a self-propelled floor cleaner. It
works using Lidar lasers or room-mapping sensors on
the vacuum to scan and map the house without human
intervention. Moreover, some vacuums are provided
with a built-in camera
This assessment examined devices from four
companies: Eufy, Ecovacs Deebot, Wyze, and
iRobot. All collected personal data and tracked the
location using the applications, and all share users'
information with third parties. Furthermore, all the
applications have the potential to invade the user's
privacy by tracking the location because they work
using Lidar lasers or room-mapping sensors on the
vacuum to scan and map the house.
The devices did not include privacy-related
information in user manual pages. Their separate
privacy policy pages were complex.
3.4 Smart Video Doorbells
Smart Video Doorbell is an internet-connected smart
device that combines a doorbell, microphone, and
camera into a single device. As a result, users can
video record who is at the door. Smart doorbells are
linked to the WIFI, and a smartphone application
permits live viewing for users.
This assessment examined Netatmo, Eufy and
Ring. Providers usually collect personal data, house
location, video recording, and face images. Further,
most devices and applications can invade the user's
privacy via cameras, microphones, and track location.
Moreover, Eufy and Ring share data with third
parties, while Netatmo applies the strictest
regulations to data collection.
Regarding privacy-related information on the
user manual page, we found that all devices do not
present any information.
3.5 Smart Displays
Smart display usually works as the hub in the smart
home and permits the user to watch YouTube, view
their photos, listen to podcasts, and maintain domestic
lighting control with their voice.
This assessment investigated Google Nest Hub,
Amazon Echo Show, and Facebook Portal. All collect
personal data, voice recording for the users and the
geolocation of the user's smartphone. Additionally,
Facebook Portal collects social contact because it
requires signing in with a WhatsApp or Facebook
account. Moreover, most smart displays share data
with third parties, including advertisers, analytics
services, and measurement partners. Besides, all
devices and their applications have the potential to
invade user privacy.
ICISSP 2023 - 9th International Conference on Information Systems Security and Privacy
594
Regarding how much privacy-related information
is presented on the user manual page, we found that
just the Facebook Portal mentioned how to turn off
the camera and the microphone.
3.6 Smart TVs
A smart TV is a television connected to the internet
with computational abilities. These upgrades to the
standard TV group allow the smart TV to operate
interactive applications and stream content from the
Internet.
This review investigated the Sony and Samsung
Smart TV. All collect personal data, and social
contact, and may share data with third parties.
Furthermore, it may share geolocation, personal
information and browsing history.
Regarding privacy-related information presented
on the user manual page, we found that Samsung
mentioned how to agree to use the privacy policy
without any further explanation. In contrast, Sony
does not present any related information.
3.7 Summary
Table 1 summarises the overall findings, denoting
whether security/privacy-related information was
Table 1: Security/privacy information at different stages.
Product
w
eb page
User
manual
FAQ
page
Other
Speaker
Echo Dot

HomePod
Nest
-

Thermostat
Amazon
-
Netatmo
Robotic Vacuum
Eufy
Ecovacs
Wyze
iRobot
Video Doorbell
Netatmo
Eufy
-
Ring
Display
Nest Hub
-
Echo Show
Portal
Smart TV
Sony
Samsung
-
identified at each stage (a dash indicates that the
element was not found). The ‘Other’ column refers to
sources such as privacy policies and notices.
4 EVALUATING FEATURE
ACCESSIBILITY
Having assessed the availability of information, the
next question was how easily a user can control
security and privacy features. Smart speakers were
selected for further examination as the most
widespread smart home device. Specifically, the
Amazon Echo, Google Nest, and Apple HomePod
devices (Figure 1) were chosen, reflecting three
popular options that typical users may own and use.
(a) (b) (c)
Figure 1: Smart Speakers selected for evaluation (a)
Amazon Echo (b) Google Nest and (c) Apple HomePod.
To explore how easily and consistently users can
locate and use relevant features, the study selected a
series of security- and privacy-relevant activities that
smart speaker users may wish to perform.
Listening and Recording: Smart speakers listen
and record user conversations to send data to
developers to enhance functionality.
Consequently, users may wish to mute the
microphone or delete these recordings.
- Microphone muting - The microphone
remains on to listen for the trigger words that
permit users to utilize the device. This
indicates that smart speakers are consistently
listening, raising privacy concern for some
users. Some devices (e.g., Google Nest and
Amazon Echo) have a physical switch to
mute or unmute their microphones.
- Deleting recordings - Captured recordings
can be deleted manually by users. Moreover,
Amazon and Google speakers enable
scheduled deletion of recordings regularly.
Audio Purchases: The smart speaker allows
users to add items to a shopping basket and
complete purchases via voice controls. As such,
smart speakers are linked to the user's payment
card and require protection to avoid unauthorised
purchases. Related protection features are:
Assessing Security and Privacy Insights for Smart Home Users
595
- Turning off audio purchase – The ultimate
protection is to disable the feature.
- Payment authorisation - Amazon Alexa
allows the user to establish a 4-digit voice
code to prevent unexpected orders or verify
purchases.
Turning off Location Services: Siri utilizes the
location of the Apple HomePod speaker to
provide regional news such as weather traffic and
nearby businesses. Moreover, location services
settings will apply to all Apple HomePods
speakers in the house. Therefore, the user should
be careful with this service because it indicates
home location, which may violate the user's
security and privacy. However, this feature is
only available via Apple HomePod.
Updating Smart Speaker Software: Updates
are released regularly to improve the device's
stability, adding new features, and closing
security gaps. Therefore, the user needs to review
the software version of the smart speaker device
and compare it to the latest version. It must be
updated if a smart speaker device runs an
outdated version.
Setting Up Voice Recognition: When a smart
speaker shares with different people, it can
convey details that are not specific to the user, or
it may share the user information with somebody
else. Therefore, this can lead to weak
authorization security issue. In cases like this, the
voice match feature will allow the smart speaker
to fetch the data specific to the user only if it
hears the user's voice.
Space constraints prevent us from discussing the
details in all cases, and so the steps involved in
deleting voice recording data are presented as a
specific example. A summary of the steps involved
across the other tasks are presented in Table 2.
Apple HomePod: The user can delete recordings
from the Home application by following the
steps in Figure 2.
While it requires fewer steps
than the others, steps at the second and third
stages are not necessarily intuitive.
Amazon Echo Dot: The user can delete
recordings manually via the Alexa application
settings, as shown in Figure 3. To delete all
recordings, they must perform at least eight
steps. Moreover, they may need help due to
ambiguous options. For example, when opening
Settings (Step 3), how will they know to choose
Alexa Privacy rather than Account Setting?
Therefore, the users may waste time trying to
find the correct route, which can be cumbersome.
Google Nest: The user can delete all records
from their Google account by following several
steps, as shown in Figure 4. It indicates that they
must select various options and scroll down via
pages to find the Voice & Audio Activity page
to delete all voice records. These steps are not
obvious due to the variety and unclear choices
and scrolling down via pages, which can make
the user try several options and undo them to
reach the chosen option. For example, if the user
opens the setting page, how will they know to
select Google Assistant and not Voice?
Figure 2: Steps required to Delete Siri History from HomePod via the Home app.
ICISSP 2023 - 9th International Conference on Information Systems Security and Privacy
596
Figure 3: Steps required to Delete all the Recording from Amazon Smart Speaker.
Table 2 summarises the overall findings and the
steps that a user needs to perform for each feature.
This highlights the difference in executing the same
feature between the three platforms.
5 CONCLUSIONS
The findings indicate that smart home devices make
it difficult for end-users to deal with security and
privacy issues in a consistent manner. In addition to
finding varying levels of information available to
guide them at the outset, users are then faced with
individual devices that perform similar functions in
varying ways, thereby complicating (and potentially
frustrating) the task of managing security and privacy
across a growing range of devices within their smart
home. This suggests a need to improve user
experience through better attention to HCI aspects.
Our resulting aim is to design and evaluate a
‘dashboard’ that consolidates status information from
multiple devices and presents it in a harmonised
manner, thereby giving the user a more readily
appreciable view of the status of their devices and the
smart home overall.
Assessing Security and Privacy Insights for Smart Home Users
597
Figure 4: Steps required to Delete all Records in Google Nest Speaker.
Table 2: Comparison of Smart Speakers Features on the selected platforms.
ICISSP 2023 - 9th International Conference on Information Systems Security and Privacy
598
REFERENCES
Alrawi, O., Lever, C., Antonakakis, M., & Monrose, F.
(2019). SoK: Security Evaluation of Home-Based IoT
Deployments. Proceedings - IEEE Symposium on
Security and Privacy, 2019-May, 1362–1380.
https://doi.org/10.1109/SP.2019.00013
Colnago, J., Feng, Y., Palanivel, T., Pearman, S., Ung, M.,
Acquisti, A., Cranor, L. F., & Sadeh, N. (2020).
Informing the Design of a Personalized Privacy
Assistant for the Internet of Things. Conference on
Human Factors in Computing Systems - Proceedings.
https://doi.org/10.1145/3313831.3376389
Emami-Naeini, P., Dixon, H., Agarwal, Y., & Cranor, L. F.
(2019). Exploring How Privacy and Security Factor
into IoT Device Purchase Behavior. Proceedings of the
2019 CHI Conference on Human Factors in Computing
Systems, 12(2019). https://doi.org/10.1145/3290605
Fabian, B., Ermakova, T., & Lentz, T. (2017). Large-scale
readability analysis of privacy policies. Proceedings -
2017 IEEE/WIC/ACM International Conference on
Web Intelligence, WI 2017, 18–25. https://doi.org/
10.1145/3106426.3106427
He, W., Golla, M., Padhi, R., Ofek, J., Dürmuth, M.,
Fernandes, E., & Ur, B. (2018). Rethinking Access
Control and Authentication for the Home Internet of
Things (IoT). In Proceedings of the 27th USENIX
Security Symposium. https://www.usenix.org/
conference/usenixsecurity18/presentation/he
Jacobsson, A., & Davidsson, P. (2015). Towards a model
of privacy and security for smart homes. IEEE World
Forum on Internet of Things, WF-IoT 2015 -
Proceedings, 727–732. https://doi.org/10.1109/WF-
IOT.2015.7389144
Kotz, D., Avancha, S., & Baxi, A. (2009). A privacy
framework for mobile health and home-care systems.
Proceedings of the ACM Conference on Computer and
Communications Security, 1–12. https://doi.org/10.11
45/1655084.1655086
Laricchia, F. (2022). Smart speakers - Statistics & Facts |
Statista. https://www.statista.com/topics/4748/smart-
speakers/
Leo Gorski, P., lo Iacono, L., Wermke, D., Stransky, C.,
Moeller, S., Acar, Y., & Fahl, S. (2018). Informal
Support Networks: an investigation into Home Data
Security Practices. www.usenix.org/conference/
soups2018/presentation/gorski
Madakam, S., & Ramaswamy, R. (2014). Smart homes
(conceptual views). 2nd International Symposium on
Computational and Business Intelligence. IEEE., 63–
66. https://ieeexplore.ieee.org/abstract/document/711
9535/
Mazwa, K., & Mazri, T. (2018). A Survey on the security
of smart homes: Issues and solutions. ACM
International Conference Proceeding Series, 81–87.
https://doi.org/10.1145/3289100.3289114
Mcreynolds, E., Hubbard, S., Lau, T., Saraf, A., Cakmak,
M., & Roesner, F. (n.d.). Toys that Listen: A Study of
Parents, Children, and Internet-Connected Toys.
Proceedings of the 2017 CHI Conference on Human
Factors in Computing Systems. https://doi.org/10.11
45/3025453
Rodden, T. A., Fischer, J. E., Pantidi, N., Bachour, K., &
Moran, S. (2013). At home with agents: Exploring
attitudes towards future smart energy infrastructures.
Conference on Human Factors in Computing Systems -
Proceedings, 1173–1182. https://doi.org/10.1145/2470
654.2466152
Satpathy, L. (2006). Smart Housing: Technology to Aid
Aging in Place - New Opportunities and Challenges.
Theses and Dissertations. https://scholarsjunction.ms
state.edu/td/3967
Udoh, E. S., & Alkharashi, A. (2017). Privacy risk
awareness and the behavior of smartwatch users: A case
study of Indiana University students. FTC 2016 -
Proceedings of Future Technologies Conference, 926–
931. https://doi.org/10.1109/FTC.2016.7821714
Worthy, P., Matthews, B., & Viller, S. (2016). Trust me:
Doubts and concerns living with the internet of things.
DIS 2016 - Proceedings of the 2016 ACM Conference
on Designing Interactive Systems: Fuse, 427–434.
https://doi.org/10.1145/2901790.2901890
Zeng, E., Roesner, F., & Allen, P. G. (2019).
Understanding and Improving Security and Privacy in
{Multi-User} Smart Homes: A Design Exploration and
{In-Home} User Study. https://www.usenix.org/
conference/usenixsecurity19/presentation/zeng
Zhang, B., Zou, Z., & Liu, M. (2011). Evaluation on
security system of internet of things based on Fuzzy-
AHP method. 2011 International Conference on E-
Business and E-Government, ICEE2011 - Proceedings,
2230–2234. https://doi.org/10.1109/ICEBEG.2011.588
1939
Zheng, S. (2017). User Perceptions of Privacy in Smart
Homes. https://dataspace.princeton.edu/handle/88435/
dsp01kd17cw477
Assessing Security and Privacy Insights for Smart Home Users
599