Journal of Hospitality Marketing & Management

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/whmm20

Artificially intelligent device use in service delivery:

a systematic review, synthesis, and research
agenda

Oscar Hengxuan Chi, Gregory Denton & Dogan Gursoy

To cite this article: Oscar Hengxuan Chi, Gregory Denton & Dogan Gursoy (2020)
Artificially intelligent device use in service delivery: a systematic review, synthesis, and
research agenda, Journal of Hospitality Marketing & Management, 29:7, 757-786, DOI:
10.1080/19368623.2020.1721394

To link to this article: https://doi.org/10.1080/19368623.2020.1721394

Published online: 11 Feb 2020.

Submit your article to this journal

Article views: 4524

View related articles

View Crossmark data

Citing articles: 57 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=whmm20
JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT
2020, VOL. 29, NO. 7, 757–786
https://doi.org/10.1080/19368623.2020.1721394

Artificially intelligent device use in service delivery: a systematic

review, synthesis, and research agenda
Oscar Hengxuan Chia, Gregory Dentona and Dogan Gursoya,b
a
School of Hospitality Business Management, Carson College of Business, Washington State University,
Pullman, Washington, USA; bSchool of Tourism and Hospitality, University of Johannesburg, Johannesburg,
South Africa

ABSTRACT KEYWORDS
This study undertakes a systematic review of Artificial Intelligence and its Artificial intelligence; review;
applications to service encounters and the hospitality industry by service; hospitality;
reviewing publications that (1) mainly discuss AI technology, (2) are in interaction; adoption
the context of services, and (3) investigate the use or the adoption of AI
technology rather than technical issues such as system design, algo-
rithms, voice recognition modules, or psychological knowledge repre-
sentations. Seven major themes are identified via a review of 63
publications. The themes are (1) current AI technology in service front-
line, (2) levels of artificial intelligence, (3) AI agents, (4) human–AI service
encounters, (5) theoretical frameworks of the acceptance of AI, (6)
reasons for adopting AI, and (7) potential challenges of AI. This study
also offers a further research agenda that highlights nine critical research
areas to guide human–AI interaction and AI adoption researches.

摘要
本研究系统地回顾了人工智能及其在服务业和酒店业中的应用，回
顾了以下出版物：（1）主要讨论人工智能技术（2）服务业（3）调
查人工智能技术的使用或采用，而不是比技术问题，如系统设计，
算法，语音识别模块，或心理知识表达. 通过对63份出版物的审查，
确定了7个主要主题. 主题是：1）当前人工智能技术在服务一线
（2）人工智能水平（3）人工智能代理（4）人类人工智能服务遭遇
（5）接受人工智能的理论框架（6）采用人工智能的理由，（7）人
工智能的潜在挑战. 本研究还提出了进一步的研究议程，重点介绍了
指导人类人工智能交互和人工智能采纳研究的九个关键研究领域.

Introduction
Artificial Intelligence (AI) consists of a collection of multiple technologies that allow
machines to sense, comprehend, act, learn (Bowen & Morosan, 2018) and exhibit aspects
of human intelligence (Huang & Rust, 2018). AI is often characterized as a “family of
technologies” which range from “low AI” applications such as self-service kiosks
(Reinders, Dabholkar, & Frambach, 2008) up to fully functional anthropomorphic robot
companions that are capable of detecting emotions and relating to humans interactively.
Although often thought of as a new technology, the theoretical framework of Artificial

CONTACT Oscar Hengxuan Chi [email protected] School of Hospitality Business Management, Carson College of
Business, Washington State University, Pullman, WA 99164
This article has been republished with minor changes. These changes do not impact the academic content of the article.
© 2020 Taylor & Francis Group, LLC
758 O. H. CHI ET AL.

Intelligence was established over 70 years ago (Bainbridge et al., 1994) and the basic
precepts have changed surprisingly little over time. Early scholars constructed artificial
neural networks in the 1950s that could “think humanly” (Russell, Norvig, & Davis, 2010,
p. 19), ushering in the era of machines that think, learn, and create. Those systems at the
time had limited “real world” applications and failed to thrive because they were limited
by then-current processing speeds, and were considered unrealistic, impractical, and hard
to use (Fluss, 2017).
Today’s generation of Artificial Intelligence may be better conceptualized as Artificial
Social Intelligence (Bainbridge et al., 1994), as it applies the intelligence techniques of
machines to social phenomena. This trend began in the 1980s when behavioral scien-
tists began conceptualizing uses for AI that could be applied in the realms of consumer
behavior and sociology. Scholars have only recently begun to focus on practical issues
related to AI device use in service delivery, the impacts of AI technology on frontline
employee–customer interactions, and the challenges related to adoption of AI devices to
enhance customer experiences (Gursoy, Chi, Lu, & Nunkoo, 2019).
With the prolific use of the AI label, the lines between automation and intelligence have
recently been blurred. Applications that are merely technological in nature but that don’t
represent actual “intelligence” are often characterized under the AI moniker even though
qualification under the conventional definitions of intelligence would require manipulation of
symbols, words, or numbers (Bainbridge et al., 1994) or a level of self-learning (Huang & Rust,
2018) rather than simple mechanical processing. Thus, as the body of research on Artificial
Intelligence is developing rapidly, it is important to undertake a review of past literature and
current practices, and to establish parameters and vocabulary to be used by researchers in
order to enable a logical, consistent progression of knowledge.
The progression of AI into the service industry creates opportunities for research to be
fractured, with multiple frameworks applied to specific, but not general, questions, which
would impair the ability of scholars to reach consensus on what is AI and an artificially
intelligent device, and how, where, when, why, and with whom artificial intelligence can or
should (or should not) be deployed. Thus, this paper aims to undertake a systematic
review of Artificial Intelligence and its applications to the service industries in general, and
the hospitality industry in particular, in order to clear confusions on what is AI and an
artificially intelligent device and to avoid the pitfalls inherent in unsystematic literature
reviews, which can include a lack of critical assessment, inappropriate conclusions, and
researcher bias (Tranfield, Denyer, & Smart, 2003). Given that Artificial Intelligence
literature in service delivery is relatively young, fragmented and divergent, and that
there is less consensus on how to use research to inform policy and practice,
a systematic review is expected to be more applicable than a traditional literature review.
Based on the results of the review, this study provides a future research agenda to guide
hospitality and service researches regarding AI service devices.

Methodology
Following the framework of Tranfield et al. (2003), the current study first identified the
inclusion and exclusion criteria for the paper collection. Then, this study conducted the
review in three stages: scoping study, comprehensive review, and thematic analysis. To
reduce human error and bias, two researchers conducted the article search individually
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 759

based on the inclusion and exclusion criteria from the mid-September to the mid-October
in 2019.

Inclusion and exclusion criteria

The current study explores the current status of AI device application and research in the
context of service transactions. Therefore, the publications reviewed in this study focus on
articles that (1) mainly discussed AI technology, (2) were in the context of services, and
(3) investigated the use or the adoption of AI technology rather than technical issues such
as system design, algorithms, voice recognition modules, or psychological knowledge
representations (e.g., Belanche et al., 2020; Kuhner et al., 2019; Lee, Kwag, & Ko, 2020).
Articles that do not meet one of these criteria were excluded by this review. For instance,
Lee et al. (2020) proposed mathematical models to optimize the operation of logistics
robot system in hotel services. Even though this study explored the AI technology in hotel
service context, the study is highly technical and not focused on service applications.
Therefore, this study was excluded.

Scoping study
An iterative “scoping study” was conducted of papers identified by a panel of researchers
and academics. Researchers first searched papers from 22 SSCI hospitality journals
(Appendix A) using key words such as artificial intelligence, smart, robot, humanoid, or
anthropomorphism to explore the current stage of AI research in the hospitality discipline.
The procedure resulted in 73 search results and an initial article pool of 15 research papers
after filtering out irrelevant articles based on the exclusion criteria. Among these 22
journals, only seven of them published AI-related articles. The Journal of Hospitality
Marketing & Management published the most AI articles (five articles). In addition, all
15 papers were published since 2017; 9 of them were published during 2019. These results
reflect that AI research is in an incubation period in the hospitality discipline.
From these papers, a snowball methodology was utilized to expand the scoping study
until an ultimate scope of 41 papers in hospitality, marketing, business, and consumer
behavior peer-reviewed journals were identified. From the scoping study, the key words
and search terms included in Appendix B were developed. Redundant terms were
excluded, and the list was refined to a final compilation of 34 unique terms.

Comprehensive review
A comprehensive review of the literature was conducted using the key words and search terms
derived from the scoping study. Since this study not only investigates the academic research
but also the application of AI technology in hospitality and service industries, to increase
thoroughness and reduce bias, the comprehensive review followed Tranfield et al. (2003)’s
recommendation and included not only papers published in peer-reviewed journals but also
unpublished studies, conference proceedings, newspapers, and trade journals from three
databases, namely ScienceDirect, Sage, and SpringerLink. Articles included in this study are
written in English and published after 2010. This stage expanded the article pool to 63
publications (Figure 1).
760 O. H. CHI ET AL.

60

50

Number of publications
40

30

20

10

0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Non-Hospitality Journals SSCI Hospitality Journals

Other Sources Total Publications

Figure 1. Publications included in this study.

Thematic analysis
In the third stage of the systematic review, a thematic analysis was undertaken to
synthesize the data. In this stage, two researchers performed a double extraction to reduce
the biases of thematic analysis (Tranfield et al., 2003). The data were extracted into eleven
background variables such as the year of publication, title, author, journal/newspaper,
service context, industry, the user of AI, country, type of study, type of AI, and major
findings. In addition, the articles were categorized into seven themes identified through
the review process. These themes are (1) Current AI Technology in Service Frontline, (2)
Levels of Artificial Intelligence, (3) AI Agents, (4) Human–AI Service Encounters, (5)
Theoretical Frameworks of the Acceptance of AI, (6) Reasons for Adopting AI, and (7)
Potential Challenges of AI. Articles were allowed to be categorized into more than one
topic.

Themes identified
Theme 1: current AI technology in service frontline
Early applications of AI were limited by a number of factors salient to the times, including
then-current technology and limited adaptability of the machines to everyday life, but also
by the limited imaginations of consumers, which largely relegated AI to the realm of
science-fiction. Contributing to the early inertia experienced with AI implementation,
predictions of the potential for Artificial Intelligence were vastly overestimated and pitfalls
to acceptance and implementation were equally underestimated. Over time, many of the
potential benefits of AI predicted in the 1950s have been largely realized (or are still in-
process) but the timetables by which technologies have been devised, implemented, and
accepted have consistently been longer than anticipated, which may serve as a warning for
contemporary prognosticators. Early predictions that a computer would become chess
champion and be able to prove complex mathematical theorems did in fact come true, but
it took 40 years rather than the projected 10 years (Russell et al., 2010, p. 21). AI-enabled
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 761

virtual bots have been available for over 25 years, but had limited success because they
were considered to be unrealistic, hard to use, and not very effective (Fluss, 2017). In this
light, current projections for 25% of the hospitality jobs in the U.S. to be automated by
2030 (PwC, 2019) or for up to 800 million jobs to be automated globally in the same
timeframe (Manyika et al., 2017) may prove equally optimistic, in timing if not in scope.
Although AI has been developing for decades and is making significant inroads, it is
still considered to be in its primitive stages, with future breakthroughs likely to eclipse the
benefits we have seen thus far. Unlike previous industrial revolutions, which focused on
agricultural and mechanical production, mass production, and brute computational
power, AI has the potential to impact the physical, digital, and biological domains in
ways that could transform virtually all aspects of society and usher in profound systemic
change (Schwab, 2016). Rather than just enhancing products, technology is now becoming
a part of the product itself (Porter & Heppelmann, 2014), coining the phrase “smart
products.” The technological advancements currently underway are characterized with
much hyperbole, alternatively being referred to as the “Fourth Industrial Revolution”
(Schwab, 2016), the “Third Wave of IT-Driven Competition” (Porter & Heppelmann,
2014), or the “Third Era of Automation” (Davenport & Kirby, 2015), among other
superlatives.
Another key differentiator of this wave is the extension of technology firmly into the
service arena and into the realms of analytical and intuitive processes vs. purely mechan-
ical or analytical tasks (Huang & Rust, 2018). Furthermore, higher processing speeds and
lower costs have enabled an unprecedented level of integration and coordination between
machines. This progression of technological prowess is now enabling the advancement of
what has been an underdeveloped industry for over half a century.

Theme 2: levels of artificial intelligence

Researchers outside of the scientific realms have diverged largely in their conceptualiza-
tion and definitions of Artificial Intelligence, and many scholars frequently use “automa-
tion” and “artificial intelligence” interchangeably. If intelligence is conceptualized as the
ability to process information, learn from experience and adapt to an environment, then
much of what is currently termed as artificial intelligence does not qualify as exhibiting
true intelligence, but is rather simply automation or the processing of large volumes of
data.
Davenport and Kirby (2015) suggest that automation occurs in three stages. First,
machines take away the “dirty and dangerous” work and relieve humans of onerous
manual labor. In the second stage, machines take over the “dull” work, and relieve humans
of routine service transactions and clerical chores. In the third and final stage, machines
take over the “decisions” and become capable of making comparable or better choices
than humans with greater speed and reliability. Under this conceptualization, AI usage in
the service industry is still transitioning from Stage 1 to Stage 2, with very little attention
yet paid to the possibilities presented by Stage 3 applications of AI to the hospitality
industry.
In studying the service industry literature, Huang and Rust (2018) specify four types of
intelligence that are largely sequential in nature: mechanical, analytical, intuitive, and empa-
thetic (Table 1). These four intelligences are useful in categorizing technological advances as
762 O. H. CHI ET AL.

Table 1. Four types of AI and their conceptualized applications.

Applications in the hospitality industry
Employee
Type of intelligence Description Customer service management
Mechanical Intelligence • Ability to perform routine and • Check-in • Screening
(“automation”) specialized tasks • Concierge/porter • Payroll
• Rule-based • Housekeeping • Administration
• Rely on a priori knowledge • Transportation
• Require little or no creativity or
“learning”
Analytical Intelligence • Ability to process information to solve • Room service/ meal • Staffing
(“weak AI”) problems or answer questions • Ordering security • Ordering
• Uses algorithms to learn iteratively but • Utility management
likely to repeat mistakes • Elevator management
Intuitive Intelligence • Ability to think creatively and adjust • Health monitoring • Training
(“strong AI”) effectively to novel situations • Guestroom comfort • Employee relations
• Learns from mistakes and changes control
future responses to maximize outcome
Empathetic Intelligence • Ability to recognize, understand, and • Diet and exercise • Employee stress
(“affective AI”) influence people’s emotions management Management
• Ability to “experience” things • Stress management
• Exhibit emotional labor/intelligence

they show an orderly progression of development and also because each stage has unique
impacts on both customer service and on employees within the service industries. Huang and
Rust also point out that the development of artificial intelligence follows a logical sequence,
starting with mechanical and progressing through analytical and intuitive up to empathetic
intelligence. Following Huang and Rust’s framework, this study summarized the current
application of AI technology in hospitality and service industries in Table 2.
An illustrative example of the four levels of intelligence applied to the same service industry
position is illustrated by the following example of a hypothetical “Barkeep 2020” system:
A patron places their beverage order by simply saying “Bartender, I’ll have my usual drink”. The
Barkeep 2020 sensors recognize the customer’s voice, process the command and a series of robotic
arms prepares the drink, stored in the system memory, and delivers it to the customer (mechanical
intelligence). The system then estimates the patron’s likely blood-alcohol level to determine whether
the customer is going to be inebriated and, projects future orders based upon the customer’s (and
other current customers’) consumption rates and prior history, anticipating inventory needs and
flagging items for the automated purchasing system to order from the vendor (analytical intelli-
gence). Based on the above-average pace of consumption for this particular customer, the system
then determines that they are in a high state of arousal and determines that social engagement might
be appropriate (intuitive intelligence). A humanoid robot then engages the customer in conversation
and ascertains that he was terminated at work. The robot sympathizes with the customer, consoles
them by enabling them to talk about it and/or take their mind off their troubles, suggests a number of
self-help sources and while also asking them whether they would like the bar to arrange for an
autonomous car to take the customer home afterwards so that they don’t make a bad day worse by
driving under the influence (emotional intelligence).

Theme 3: AI agents
Artificial Intelligence Agents are considered to have two dimensions – presence and
embodiment (Tung & Law, 2017). Presence refers to whether or not the agent has
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 763

Table 2. Current applications of AI technology in hospitality and service industries.

Intelligence Study Application
Mechanical Intelligence Halzack (2017); Humanoid robot (Pepper) interacts with customers
Mende et al. (2019) in hotels, cruise ships, and Nescafe stores
Johnson (2016) Self-service kiosks at fast-food restaurants
Hennessy (2017) Robots make pizzas at Zume in Silicon Valley
Bowen and Morosan (2018) Caliburger robot cooks hamburgers
Pinillos et al. (2016) Sacarino interactive bellboy robot walks alongside
Rodriguez-Lizundia et al. (2015) guests, provides information on local and hotel-
related services
Crook (2014) Boltr robotic butler at Aloft delivers linens and
amenities to guest rooms
Golden (2014) Robot arm bartenders make drinks at “bionic bar”
on Royal Caribbean cruises
Hertzfeld (2019) Hotel Henn Na in Nagasaki operates with 243
robotic staff
Hospitality Technology (2019a) Pactum software manages vendor accounts and
negotiations to increase contracting efficiency
Hospitality Technology (2019b) Vinpearl Nha Trang resort uses facial recognition to
facilitate guest check-in, security, and provide resort
access
Phillips (2018) Creator restaurant in San Francisco serves burgers
made completely by robots
Kahn (2017); Victor (2014) Dominos’ DRU autonomous delivery vehicle
Analytical Intelligence Trejos (2016) Connie robot concierge at Hilton McLean answers
guest questions and gives directions
Ross (2019) Rose chatbot at Cosmopolitan of Las Vegas fulfills
guest request and answers 82% of inquiries without
human intervention
Hospitality Technology (2019c) RLH Corporation employs an AI Virtual Agent for
Central Reservations at 1400 system hotels
Hospitality Technology (2018) AutoClerk PMS system enables hotel staff and
guests to communicate via texts to order services,
report issues, and generate maintenance work
orders
Foster et al. (2017) Software to enable robot bartenders to estimate the
social states of humans
Intuitive Intelligence Edelman and Singer (2015) Hotel smartphone apps that can check guests into
room, act as room key, and control guestroom
environment, then make recommendations for
dining/entertainment
Poggi (2017) Chatbots used to drive online sales at fast-food
restaurants
Gursoy et al. (2019) Spencer virtual agent assists KLM airline customers
to enhance travel experience
Ross (2019) AllyO recruiting software locates potential job
candidates and induces them to apply, then screens
candidates for overall suitability
Hospitality Technology (2019d) Zoe personal cruise assistant on MSC Cruises
provides information, makes bookings, and interact
with guests
Hospitality Technology (2019e) Wave RMS revenue management system optimizes
hotel revenue management strategies
Javelosa (2017) BlackRock plans to replace money managers with
computer algorithms
Williams (2018) Facial recognition technology can be used to check
hotel guests into their rooms and accept payment
automatically, assess their sentiments, and track
traffic patterns
Empathetic Intelligence Huet, (2016) Replika personalized frontline communication
Stuart (2016) Emotion-aware digital devices can detect
consumer’s emotions and react by changing the
room environment, ordering coffee, etc.
764 O. H. CHI ET AL.

a virtual presence (e.g., Siri) or a physical embodiment that is visible to the guests.
Embodiment can take multiple forms, ranging from fully co-present service robots to
digital images (for example, on a computer monitor). Based on this concept, the AI
devices used in modern service transactions can be categorized into four types of agents.

Smart devices
Smart technology has the “ability to sense changes in their circumstances and execute
measures to enhance their functionality under the new circumstances” (Worden, Bullough,
& Haywood, 2003, p. 1). For example, a smart thermostat (e.g., Ecobee4) can detect the
changes of room temperature and users’ preferences and provide automatic personalized
temperature control. Some other commonly seen smart products include smart lights (e.g.,
Philips Hue), smart security cameras (e.g., NetGear Arlo Q), or smart speakers (e.g., Amazon
Echo). Many of these products look similar to conventional products but have AI systems
embedded in the product to sense the environment and respond accordingly. Therefore,
most of the smart devices are not virtually present nor physically embodies.
Smart products have been applied to hotel services in a variety of capacities. Hotels that
adopt a series of these smart technology products are smart hotels or intelligent hotels
(Wu & Cheng, 2018). Smart hotels enable guests to control and personalize their rooms
through voice commands or their smartphones or other devices and provide guests
a convenient lodging experience.

Self-service technologies
Self-service technologies (SST) are interfaces that enable customers to produce a service
independent of direct service employee engagement. The three most popular SST plat-
forms are vending machines, self-service kiosks, and web applications. The use of self-
service technologies has increased dramatically in recent years and is projected to continue
to experience significant growth, increasing from $54.4B in 2016 to $83.5B by 2021 (Lee &
Cranage, 2019). SST applications are embraced by companies due to their cost-
effectiveness and scalability, but implementation of SSTs involves potential friction in
that it requires training of customers (Lu, Cai, & Gursoy, 2019) and some customers react
negatively to having the option of human interaction taken away (Reinders et al., 2008).
While most of the SSTs provide automated services, AI applications are becoming an
important component of these SSTs in recent years.

Chatbots
Chatbots are another type of AI agent that can effectively communicate with users. Chatbots
have been used in different service contexts such as medical consulting (Nadarzynski, Miles,
Cowie, & Ridge, 2019), tourism recommendation (Ukpabi, Aslam, & Karjaluoto, 2019), or
hotel booking (Parmar, Meshram, Parmar, Patel, & Desai, 2019). These devices process
humans’ language and respond to users using simulated human language. Thus, these devices
are not embodied but they are considered to maintain a virtual presence. Chatbots are most
commonly associated with online or telephonic transactions (Byrne, 2018). Due to the
advanced anthropomorphism technology, the customer may not even be aware that they
are speaking with an AI device (Hyken, 2017). Chatbots offer advantages over human service
agents in that they can process virtually unlimited volumes of transactions simultaneously,
store vast reservoirs of information, and are less prone to errors or emotional variability or
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 765

fatigue (Byrne, 2018; Robinson et al., 2019). Conversely, Chatbots have two major disadvan-
tages. First, an ethical issue is raised when the highly humanlike chatbots lead people to believe
that they are interacting with a human being while they are in fact interacting with a machine
(Robinson et al., 2019). Second, when a company utilizes chatbots rather than human
employees to interact with customers, the customers may feel undervalued by the company
(Byrne, 2018).

Service robots
Service robots have received a considerable amount of attention in AI literature, although
discussion diverges on a number of key dimensions of form and function. Although
service robots are uniformly considered to be physically embodied agents, scholars
differentiate whether service robots are necessarily anthropomorphic, whether they
require social skills, and to what level they embody artificial intelligence.
Service robots have multiple definitions in the literature, including the IFR definition of
service robots as “a robot that performs useful tasks for humans or equipment excluding
industrial automation application” (IFR, 2016, p. 9). Bowen and Morosan (2018) define
service robots as “physically embodied artificially intelligent agents that can take actions that
have effects on the physical world.” Wirtz et al. (2018) define service robots as “system-based
autonomous and adaptable interfaces that interact, communicate, and deliver services to an
organization’s customers.” Examples of service robots employed in the hospitality industry
include Sacarino, an interactive bellboy that moves alongside guests, provides minimal basic
services, and can provide scripted information about the city (Pinillos, Marcos, Feliz, Zalama,
& Gomez-Garcia-Bermejo, 2016). Other forms of service robots, however, do not interact
with guests and simply perform programmed repetitive tasks, such as the Caliburger robot
that cooks hamburgers and places them on trays (Bowen & Morosan, 2018), or the robot arm
bartenders at Royal Caribbean cruise lines’ “Bionic Bar” (Tung & Law, 2017).
Social robots are a subset of service robots and have the added capability of interacting
and communicating with humans and following behavioral norms (Rodriguez-Lizundia,
Marcos, Zalama, Gómez-García-Bermejo, & Gordaliza, 2015). Transfusing social skills into
robots is one of the more profound advances in AI over recent years and has enabled AI to
make significant inroads into the service industry. It is important to distinguish between
service robots that provide guest services and service robots that assist employees, as much of
the potential benefits and more valuable applications of AI may accrue “behind the scenes”
and outside of the purview of customers. Social robots can be autonomous or semi-
autonomous and can take a variety of forms of embodiment. Tangible actions (delivering
a package, giving a haircut, etc.) will obviously require tangible robots, whereas intangible
actions could be performed by virtual robots, videos, or holographs (Wirtz et al., 2018).
Humanoid service robots are believed to be capable of more meaningfully engaging
customers on a social level (Mende, Scott, van Doorn, Grewal, & Shanks, 2019). The
majority of recent literature discusses service robots as anthropomorphic machines, and
a number of hotel companies have deployed robot bellboys, concierges, and front desk
agents. The Boltr robotic butler delivers towels and amenities to guestrooms (Crook,
2014), the Henn-na Hotel in Japan that deployed 243 robot agents (Lewis-Kraus, 2016),
and the Pepper robots deployed worldwide (Mende et al., 2019) are examples of service
robots in current usage.
766 O. H. CHI ET AL.

Although anthropomorphic service robots are among the most dramatic evolutions in
the service realm, they are also the source of a considerable amount of the resistance
experienced by customers (discussed later).

Theme 4: human–AI service encounters

The development of AI robots with social skills enables the interaction between humans
and robots (Wirtz et al., 2018) as a counterpart of frontline service interactions. These
interactions lead users to perceive the robot as a social entity (Van Doorn et al., 2017).
Therefore, potential issues caused by this special social entity are not only associated with
the performance of the device or the sheer logistics involved in coordinating hardware,
software, cloud, and other technologies but also related to consumer behaviors and human
psychological challenges (Rizzuto & Reeves, 2007). For this reason, the interaction
between human and AI robots is another crucial issue in the context of services (Kunz,
Kristina, Jos, & Ben, 2018) and has attracted many researchers’ attention.
This part of review was organized using Zhang and Li's (2004, 2005) framework of
human–computer interaction. This framework originally focused on assessing the litera-
ture regarding human–computer interaction in management information system domain
(Zhang & Li, 2004). Later studies have utilized this framework to evaluate AI–human
interaction studies (e.g., Rzepka & Berger, 2018). Zhang and Li’s framework suggests that
studies of human–computer interaction can be categorized into five segments: system,
user, task and context, interaction, and outcome. The first three segments refer to those
studies which investigate the characteristics of the computer system (e.g., system complex-
ity or hedonic features), of the users (e.g., personal traits or technology readiness), and/or
of the context (e.g., entertainment or office job). The interaction segment includes studies
that examine human’s perception of a system (e.g., perceived usefulness), their attitudes or
intentions toward the system (e.g., intention to use), or their behaviors related with the
system (e.g., purchasing the system). The outcome segment refers to the final result from
using the computer system, such as customers’ perception of the service quality or service
providers’ financial outcomes. According to Zhang and Li (2004), system, user, task and
context influence interaction, which further influences the overall outcomes.

Characteristics of AI devices
The interaction between AI employees and human customers is the most common type of
human–AI interactions in service deliveries (Anderson, Rainie, & Luchsinger, 2018). In this
service context, research regarding the characteristics of AI devices can be categorized into
three groups: anthropomorphism, the behavior of robots, and functionality of devices.
Anthropomorphism refers to the humanlike feature of AI devices (Gursoy et al., 2019).
Previous studies have found that anthropomorphism of service devices can have mixed impacts
on customers’ attitude and behavior. For example, in the setting of hotel services, anthropo-
morphism has been found to negatively impact customers’ perception of robots (Yu, 2019) and
decrease customers’ willingness to use AI devices (Lu et al., 2019) in both full-service and
limited-service hotels (Lin, Chi, & Gursoy, 2019). Scholars argue that highly humanlike devices
may threaten user’s human-identity (Fernández-Llamas, Conde, Rodríguez-Lera, Rodríguez-
Sedano, & García, 2018; Gursoy et al., 2019), leading to a negative evaluation toward the use of
these devices. However, in lab experiments, the anthropomorphic features of an AI device have
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 767

been shown to positively impact customers’ trust (van Pinxteren, Wetzels, Rüger, Pluymaekers,
& Wetzels, 2019) and intention to build rapport with these devices (Qiu, Li, Shu, & Bai, 2019).
Compared with machine-like voices, customers show more trust toward the devices with
humanlike voices (Xu, 2019). Furthermore, humanlike gestures can promote customers’ interest
in interacting with these devices (Xu, 2019). These mixed findings emphasize that more
comprehensive studies are necessary to explore customers’ responses to different levels of
anthropomorphism in service contexts.
Behaviors of an AI device, such as language and motions, also critically influence service
interactions. A field study found that the direct speech (e.g., directly talking to guests) used by
a robot is more effective in attracting guests’ interactions than indirect speech (e.g., a robot
talks to another robot) (Pan, Okada, Uchiyama, & Suzuki, 2015). Furthermore, in the context
of healthcare, an experimental study found that the polite speech used by a service robot may
increase users’ positive reactions (e.g., satisfaction) but will decrease users’ perceived benefit of
compliance (Lee, Kim, Kim, & Kwon, 2017). Studies also found that different motion modules
used by AI devices may result in different users’ emotional reactions. For example, in a study
that investigated the emotional impact of service robot behavior, researchers found that
singing and dancing behaviors of a social robot effectively evoke autistic children’s emotional
responses (Aziz, Moganan, Ismail, & Lokman, 2015). Researchers also found that social robots
which look active (vs. in sleep mode) are more likely to attract customers to interact with them
(Rodriguez-Lizundia et al., 2015). However, a subsequent study suggests that the impact of
behaviors of an AI device on customers interaction may be moderated by the level of
anthropomorphism. An experimental study conducted in the hotel service context found
that literal language used by an AI humanoid robot increases customers’ evaluation toward
the robotic service (Choi, Liu, & Mattila, 2019). However, this effect is not significant if the AI
device is a service kiosk. Thus, researchers proposed that when an AI device is humanlike,
customers may evaluate this device in the similar way that they evaluate human employees
(Choi et al., 2019).
Functionality is a major determinant in customers’ evaluation toward AI devices.
A previous study found that robotic embodiment and human-oriented perceptions are
positively related to customer lodging satisfaction since they provide customers with a novel
service experience (Tung and Au, 2018). This finding suggests that customers prefer AI
devices to be physically visible and equipped with intelligence technology that enables them
to effectively interact with humans. Furthermore, other research suggested that users desire
personalized features of AI service devices. A study that explored the preferred features of
service robots by different stakeholders in an eldercare context found that all the user groups
rated the ability of customization as a highly demanded robot feature (Bedaf, Marti, & De
Witte, 2019). While customers demand the performance of AI devices, they also indicate
concern over whether AI devices are under their control, and that availability of interruption
options that enable customers to terminate the running of AI devices can increase customers’
perceived behavioral control of using the robot, leading to a higher level of customer
evaluation toward these devices (Jörling, Böhm, & Paluch, 2019).

Characteristics of users
Several existing studies suggest that different customers may have different attitudes and
intentions to interact with AI devices due to their level of technological knowledge, IT
skills, and concern over cyber-security among the factors influencing their intention to use
768 O. H. CHI ET AL.

an AI chatbot (Nadarzynski et al., 2019). Another study found that customers’ level of
technology attachment (e.g., technology dependence, technology identity, technology
affection, and technology social bonding) increases their satisfaction with robotic services
and trust toward AI service devices in the context of hotel services (Wu & Cheng, 2018).
Furthermore, robot use self-efficacy has been found to be positively associated with users’
functionally and socially acceptance of service robots (Latikka, Turja, & Oksanen, 2019).
Other research has identified that customers’ hedonic motivation and their social influence
are major factors influencing their evaluation of AI service devices in hotel services. Lin et al.
(2019) investigated the antecedents of hotel customers’ willingness to use AI service robots in
different hotel service contexts. They found that, compared with other predictors (social
influence, anthropomorphism, performance expectancy, and effort expectancy), customers’
hedonic motivation is the most salient factor that drives their emotion toward AI robotic
services. A similar result was also found by a qualitative study, which argued that the main
motivating factors leading Taiwanese customers to use AI robot in hotels are to seek fun and
satisfy curiosity (Kuo, Chen, & Tseng, 2017). Furthermore, previous studies highlighted the
impact of hotel customers’ social influence on their behavior to adopt AI service devices
(Gursoy et al., 2019; Lin et al., 2019). These studies found that customers who have a high
perceived social influence of using AI service devices are likely to give the use of these devices
a positive evaluation (Gursoy et al., 2019). Lin et al. (2019) further found that perceived social
influence plays a more important role when the purchase decision is perceived as being less
important (vs. highly important) by customers.
The demographic profile of customers also influences their attitude toward AI robotic
services. Several previous studies found that males generally have a more positive attitude
toward service robot in the context of hotel services (Ivanov, Webster, & Garenko, 2018)
and health care (Hudson, Orviska, & Hunady, 2017). In addition to gender, these studies
also found that people who live in more developed area (Ivanov et al., 2018), who are
younger (Hudson et al., 2017), and who have higher education background (Hudson et al.,
2017) are more likely to accept robotic services.

Characteristics of tasks and contexts

When customers are evaluating the use of AI robots, they are likely to make the evaluation
under certain contexts. For example, Wu and Cheng (2018) found that perceived risk may
moderate customers’ satisfaction of smart hotel devices and their intention to sharing this
lodging experience on the social media. When the customer perceives the risk associated
with using smart hotel devices is high, even though they are satisfied with the services
provided by these smart hotel devices, they are less likely to share the experience with their
friends. Lu et al. (2019) found that customers’ perceived facilitating condition positively
relates to their willingness to use AI robotic devices in hotel service transactions. The
result of their study revealed that the organizational or/and technological support from
hotels or robot designers can mitigate customers’ perceived challenges associated with
using robotic services and can minimize barriers for customers to accept the use of AI
service devices.
Robot-service fit is another critical factor that influences customers’ perception. Ivanov
et al. (2018) found that Russian adults are more likely to accept some robotic hotel services
while rejecting others. For example, they are more receptive toward the robots that
transport goods (e.g., carrying luggage or linens), provide information (e.g., offering
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 769

information regarding local tourism attractions), or process payments, but object to the
use of robots to provide security or jobs requiring skin contact (e.g., massage).
Furthermore, Chan and Tung (2019) conducted an experimental study to investigate the
moderation effect of hotel segment (budget, midscale, or luxury hotel) on the relationship
between robotic service and customers’ evaluation of lodging experience. They found that
robotic services cause more impact in the contexts of budget and midscale hotels than
luxury hotels, since the service value provided by the service robot is not significantly
valued by luxury hotel customers. Lin et al. (2019) investigated customers’ acceptance of
AI robotic services in full-service hotels and limited-service hotels. Their results suggested
that customers’ behavioral intentions are generated differently across these two service
contexts. More specifically, they found that when evaluating robotic service, full-service
hotel customers depend less on social influence, their overall emotion toward robotic
services is not influenced by effort expectancy, and their objections are less likely to be
influenced by their emotion toward the robot. These evidences suggest that customers’
demands for interaction with human employees are more stable in full-service hotels than
in limited service hotels.

Interaction
According to Zhang and Li (2004), human–computer interaction framework, perception
of AI devices is one of the major research opportunities in the human–AI interaction
process. Recent studies have investigated customers’ performance expectancy (Gursoy
et al., 2019; Lin et al., 2019; Lu et al., 2019) and effort expectancy (Gursoy et al., 2019;
Lin et al., 2019). These studies suggest that, while AI service devices can provide accurate
and consistent services, interacting with these AI devices in service transactions requires
customers to emotionally accept the interaction with machines rather than human
employees. Thus, customers may need to adjust their psychological states in order to
participate in this AI–human interaction. Previous research also investigated customers’
perceived responsibility of service robots for service outcomes (Jörling et al., 2019). This
study found that users perceive a low responsibility of service robots for positive out-
comes. That is, in the context of a successful service outcome (e.g., cleaned floor), users are
more likely to give credit to a manual service product (e.g., standard vacuum cleaner) than
a service robot (e.g., robot vacuum cleaner). This study further found that this is caused by
the autonomous features of service robots decreasing users’ perceived behavioral controls.
In addition, through in-depth phenomenographic interviews, Čaić, Odekerken-Schröder,
and Mahr (2018) identified six perceived roles of social service robots in the context of
eldercare: “enablers, intruders, allies, replacements, extended selves, or deactivators”.
Different perceived roles may significantly influence a user’s interaction with service
robots. Thus, it is important to conduct both quantitative and qualitative studies to
investigate customers’ perceived roles of service robots in hospitality service contexts.
Other studies also investigated customers’ trust toward AI devices (Lee, Lin, & Shih,
2018b; Nadarzynski et al., 2019; van Pinxteren et al., 2019; Wu & Cheng, 2018; Xu, 2019),
and rapport building with robots (Qiu et al., 2019). For example, Wu and Cheng (2018)
found that customers’ technology attachment and satisfaction of using these devices
increase their trust in smart hotel devices. Furthermore, van Pinxteren et al. (2019)
found that anthropomorphism drives trust, willingness to use, and enjoyment. Qiu
et al., (2019) investigated customers’ rapport building with robot employees. Their
770 O. H. CHI ET AL.

findings suggest that while customers build rapport with human hotel employees, they
also tend to build rapport with humanlike robots.
Several studies concentrated on investigating customers’ willingness to accept the use of
AI devices (Baisch et al., 2017; Conti, Di Nuovo, Buono, & Di Nuovo, 2017; Gursoy et al.,
2019; Hebesberger, Koertner, Gisinger, & Pripfl, 2017; Kuo et al., 2017; Latikka et al., 2019;
Lin et al., 2019; Lu et al., 2019; van Pinxteren et al., 2019; Yu, 2019). Some of these studies
explored the antecedents of willingness to use, such as anthropomorphism (e.g., Lu et al.,
2019; Yu, 2019) or self-efficacy (Latikka et al., 2019); Other studies investigated customers’
tendency to accept the use of AI service devices in different service contexts (e.g., Conti
et al., 2017; Hebesberger et al., 2017). These studies highlight that customers’ willingness
to use AI service devices is influenced by a complex set of factors, including the char-
acteristics of users, devices, and contexts, and that this acceptance is further influenced by
customers’ perceptions and other factors (e.g., emotion) generated during the interaction
process.
To explore the complex nature of customers’ acceptance of AI service device in hotel
services, Lin et al. (2019) used the Artificially Intelligent Device Use Acceptance (AIDUA)
framework proposed by Gursoy et al. (2019) to investigate customers’ willingness to accept
or object to the use of AI service robots. They found that customers’ acceptance is
influenced by perceived social influence, hedonic motivation, perceived anthropomorph-
ism of the robot, performance expectancy, effort expectancy, and their overall emotion
toward the robotic service. In addition, their study confirms that in service transactions,
the acceptance of uses is likely to be selective and multidimensional. That is, even though
customers are willing to use robotic services due to the utility performance and the novelty
value, they refuse to use them due to the need for human–human interaction. Their
research highlighted customers’ mixed evaluation toward hotel robotic services.

Theme 5: theoretical frameworks of the acceptance of AI

Many existing studies related to the acceptance of AI devices in the service contexts utilized
traditional technology acceptance theories. For example, Conti et al. (2017)’s study, which
investigated practitioners’ attitudes toward the use of AI service robots in the education of
children used the Unified Theory of Acceptance and Use of Technology. However, scholars
argued that technology acceptance theories were developed to explain the acceptance of
traditional non-intelligence technology products (Gursoy et al., 2019). Whereas the intelligent
nature of AI service devices leads some predictors proposed by these traditional acceptance
theories to be irrelevant in the context of AI devices (Lu et al., 2019). Furthermore, AI devices
used in the service context are designed to interact with people as social entities (Van Doorn
et al., 2017). Therefore, these devices may also influence customers emotionally or psycholo-
gically during service transactions (Van Doorn et al., 2017).
Wirtz et al. (2018) suggest that customers’ acceptance of service robots can be influenced by
factional, relational, and social-emotional elements. The functional element (e.g., perceived
ease to use, perceived usefulness, and subjective social norms) focuses on customers’ perceived
utility of the robot. The relational element (e.g., trust and rapport) focuses on customers’
perceived relationship with the robot. Lastly, the social-emotional element (e.g., perceived
humanness, perceived social interactivity, perceived social presence) focuses on customers’
psychological evaluation of the robot as a social presence that participates their social life.
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 771

Gursoy et al. (2019) developed and validated the Artificially Intelligent (AI) Device Use
Acceptance theory that explains the interrelationship among the antecedents of acceptance
in service transactions. The theory observes that customers’ acceptance behaviors are
generated through a three-stage process: primary appraisal, secondary appraisal, and
outcome stage. In the primary appraisal stage, customers create an initial attitude toward
AI devices driven by their social influence, hedonic motivation, and perception of the
anthropomorphism of an AI device. In the secondary stage, customers evaluate the AI
device based on performance and effort expectancy and generate an emotion toward the
device accordingly. In the outcome stage, customers’ acceptance of or objections to the use
are driven by their emotions. This framework was confirmed by Lin et al. (2019) in the
context of full-service hotels and limited-service hotels.
As discussed, anthropomorphism of an AI service device plays a significant role in
customers’ acceptance to use and attitude toward AI devices. Of the few existing studies
that concentrated on exploring the impact of anthropomorphism, a number of them draw
on the Uncanny Valley Theory (e.g., Qiu et al., 2019; Yu, 2019) developed by Mori in
1970. The Uncanny Valley Theory predicts the relationship between an object’s level of
human likeness and people’s emotional response to the object, suggesting a non-linear
relationship between these two variables (Mori, 1970). More specifically, as the level of
anthropomorphism increases, people’s positive emotion toward the anthropomorphic
object increases until the anthropomorphism reaches a certain level, above which people
perceive that the anthropomorphic object threatens their human identity. Thus, their
emotion toward the object drops dramatically. However, once the anthropomorphic object
achieves real human attributes, people’s emotional reaction becomes increasingly positive
again due to the human-to-human empathy.
The Uncanny Valley Theory was further expanded by Murphy et al. in 2019. Concentrating
on hospitality and tourism services, they suggested that anthropomorphism can impact
human–AI robot interactions via three possible outcomes: anthropomorphic loyalty, accep-
tance, and affective reaction. According to Murphy, Gretzel, and Pesonen (2019), if
a customer can create a loyalty toward an inanimate brand, he/she can also create a loyalty
toward an anthropomorphic robot. However, customers may show very negligible loyalty to
a humanlike robot until the level of anthropomorphism reaches a very high level.
Furthermore, Murphy et al. argue that customers may accept AI service robots at a certain
level as long as they do not perceive that the robots threaten their human identity.

Theme 6: reasons for adopting AI

In the context of services, there is an increasing trend toward using AI robots to replace
human employees rather than complementing them, as evidenced by robot bellboys
deployed in hotels (Gursoy et al., 2019), robotic bartenders (Tung & Law, 2017) or by
the recent (and temporary) development of Henn-Na hotel with 243 humanoid robot
workers (Hertzfeld, 2019). This trend toward robotic service agents appears to be driven
largely by economic factors such as labor cost-saving (Gursoy et al., 2019; Manyika et al.,
2013; West, Clifford, & Atkinson, 2018) and/or social factors such as an aging society
(Kuo et al., 2017; Lee et al., 2018a) or the movement of raising minimum wages (Mcafee &
Brynjolfsson, 2016).
772 O. H. CHI ET AL.

In addition to the financial and social factors, the adoption of AI technology also
provides considerable service-enhancement opportunities (Marinova, de Ruyter, Huang,
Meuter, & Challagalla, 2017; Qiu et al., 2019), which are often overlooked in the “auto-
mation vs. jobs” debate. For example, Prentice, Dominique Lopes, and Wang (2019)
explored the impact of employees’ emotional intelligence and artificial intelligence of
their working environment on their job performance. They found that even though
emotional intelligence predicts job performance, the artificial intelligence of the working
environment can effectively compensate employees who have low emotional intelligence,
leading to a higher job performance. In the context of health-care services, Lee and
Cranage (2019) found that service robots can reduce nurses’ workload by taking over
the nonvalue-added nursing activities such as transporting linens or cleaning. By doing so,
these robots enable nurses to provide more care and better services to patients.
Furthermore, AI service devices enable services to be delivered at a lower cost, allowing
services which were cost prohibitive to be offered to customers resulting in greater overall
service value (Bowen & Morosan, 2018)
The services provided by AI devices can also provide additional and special service
value. Previous studies found that robotic services (vs. services provided by human
employee) are more highly associated with sensory and intellectual experiences (Chan &
Tung, 2019). Other studies also suggest that in hospitality services, the interaction between
AI robots and customers cocreates novel experiences (Van Doorn et al., 2017; Tung & Au,
2018). Some customers even seek to interact with service robots and find opportunities to
build rapport with them (Tung & Au, 2018).
Several existing studies also revealed that adopting AI robots in services can also indirectly
benefit the overall service outcome. For instance, Tussyadiah and Miller (2019) found that
a robot can act as an agent to influence customers’ behaviors that are not directly associated
with the robot. Utilizing the theory of the watching-eyes effect, their study suggests that when
a robot employee is present, hotel customers are more likely to exhibit pro-environmental
behavior. Furthermore, Pan et al. (2015) found that while two or more customers are inter-
acting with a service robot simultaneously, the customers are more likely to interact with other
customers. From this perspective, the use of service robots may promote customers’ social
interaction in the context of hotel services. Mende et al.’s (2019) study focused on customers’
compensatory responses evoked by robotic services in a restaurant setting. They hypothesized
that humanlike robots may result in customer discomfort since these robots threaten their
human identity. This discomfort further triggers customers’ compensatory responses such as
food craving. The results of their study confirmed this hypothesis, indicating that customers
who are served by humanlike robots (vs. by human employees) are likely to order more food.

Theme 7: potential challenges of AI

Perceived identity threat
In order for AI service devices to be truly accepted in service transactions, they should not only
perform the intended functions but also need to be socially acceptable (Rodriguez-Lizundia
et al., 2015). However, as discussed herein, studies have found that a majority of people who
prefer a robot companion aim to achieve utility benefits (e.g., assistant) rather than psycho-
logical benefits (e.g., friendship) (Rodriguez-Lizundia et al., 2015). According to the Uncanny
Valley theory, people may socially accept an object as their companion only when the
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 773

anthropomorphism of this object reaches a very high level (e.g., humanlike empathy AI),
which will result in generation of empathy toward this object and accept this object as a social
entity. However, the current technology is not sufficient to enable AI devices to achieve this
level of humanness. Thus, in the current stage of AI development, people tend to only accept
AI service devices for their utility benefit and perceive that these devices threaten their human
identities (Gursoy et al., 2019).

Network security and information privacy

Concern about network security and information privacy is another challenge of AI
adoption (Wirtz et al., 2018). Data collection, storage, and access are the fundamental
functions that enable AI devices to process environmental changes (e.g., talk to different
customers) and solve problems. The current technology allows these devices to record
comprehensive data from a customer such as appearance, voices, behaviors, backgrounds,
or habits. These data are then potentially shared across a broad spectrum of devices, at
which point customers have limited ability to control their personal information or
protect themselves against security risks (Denning, Matuszek, Koscher, Smith, & Kohno,
2009; Holder, Khurana, Harrison, & Jacobs, 2016)

Resistance to change – customers

Using AI service devices to deliver services that were used to be delivered by human
employees can challenge customers’ perceptions of services (Ackerman, 2016) because
customers may feel personally devalued by being forced to intact with an AI device. In the
context of hospitality services, the quality of social interactions included in the service
provision significantly influences customers’ perceived service enjoyment (Dedeoglu,
Bilgihan, Ye, Buonincontri, & Okumus, 2018). Thus, customers may potentially devalue
or outright refuse using services provided by AI devices due to their preference for social
interaction (Gursoy et al., 2019; Lin et al., 2019).

Resistance to change – employees

In the context of AI services, the human employee is also a role player in service
encounters (Robinson et al., 2019). Therefore, AI devices which are used to provide
services not only interact with customers but also with employees. In fact, implementing
AI robots as service employees involves significant considerations in not just technology,
costs, or service quality, but also the acceptance and attitudes of human employees.
Bedaf et al. (2019) explored the preferred characteristics of service robots by different
stakeholders in an elder care context. This study found that the users of service robots
prefer a highly customizable robot that can match individuals’ needs. Furthermore, Yoon
and Lee (2019) developed a framework that focuses on the technology side of service
robots in the healthcare industry. Through a systematic literature review, their study
identified 10 key features of service robots (user friendly, interaction with human, easy
access, fast response, problem diagnosis and solution, accuracy, assurance, reliability,
transaction verification and security, and remote monitoring) that are demanded by
both customers and employees. All these studies emphasized that the adoption of service
robots cannot be successful without considering the needs and benefits from both custo-
mers and employees.
774 O. H. CHI ET AL.

However, existing studies reveal that employees generally have more negative attitudes
toward AI service devices than positive attitudes. Li, Bonn, and Ye (2019) investigated the
impact of hotel employee’s awareness of AI and robots on their turnover intention and
found that AI and robotic awareness is positively related to employees’ turnover intention
due to the threat of being replaced. They also found that perceived organizational support
weakens the relationship while competitive psychological climate enhances the relation-
ship. Lee and Cranage (2019) explored Korean caregivers’ perceptions of service robots in
nursing care services. Even though participants confirmed that service robots can assist
them on their routine tasks, they express concern that these robots will interrupt the
relationship between human-caregivers and patients.
In addition, Hebesberger et al. (2017) explored employees’ social acceptance of the use
of service robots as their coworkers in health-care services and found acceptance of service
robots to be indecisive. These caregivers are found to accept the use of social robots to
assist them in performing their responsibilities and duties, but do not want to share their
workspace with robots. In the education context, Conti et al. (2017) explored the accep-
tance of using service robots to educate and take care of children with developmental
disabilities and revealed that experienced practitioners held much more negative attitudes
toward the use of service robot than did student participants.

Employment disruption
One of the more impactful themes in AI literature is the predictions for and implications
of technology replacing human labor and making humans expendable. McKinsey’s pro-
jection for 800 million jobs to be automated globally by 2030 (Manyika et al., 2017) may
seem optimistic, but even a fraction of that level of displacement will cause significant
disruption in the global economy, with both positive and negative implications for
industry. The capital investment required to introduce such a large number of social
service robots, the ensuing cost savings from automation and the effect of such displace-
ment of human employees would dramatically reshape entire industries, which would be
further affected by the evolution that would likely occur from the growth in idle human
labor.
Davenport and Kirby (2015) warn that unless new tasks are found to replace the functions
that automation takes away, the psychological and social ills of unemployment will expand,
and the disruption will be particularly profound as AI takes over “knowledge work.” In past
industrial revolutions, similar concerns were expressed over labor disruption; however,
although short-term disruption did occur in select industries or locations, the overall long-
term impacts were to increase the need for labor in other sectors of the economy.
Furthermore, past revolutions threatened onerous manual labor tasks but did not infringe
on or threaten human intelligence functions (Davenport & Kirby, 2015). In the current, post-
industrial era, labor has shifted significantly from manufacturing to service tasks because
service requires heterogeneous contextual interactions which mechanical AI robots could not
replicate (Huang & Rust, 2018). As AI progresses through Huang and Rust (2018) four levels
of intelligence, the key for human employees to remain “important” and “relevant” will
depend on their ability to evolve their skills and knowledge in enhancing customers' service
experiences by most effectively managing the utilization of AI-powered service devices to
provide enhanced customer experiences as the AI progresses.
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 775

As AI transitions into social and analytical functions, the service sector will find itself at an
inflection point similar to the industrial revolution in manufacturing (Wirtz et al., 2018). In
the hospitality and service industries, the level of human labor disruption has the potential to
be even more dramatic, with Price Waterhouse Coopers forecasting 25% of the current
hospitality jobs in the US to become automated by 2030 (Bowen & Morosan, 2018).
Frey and Osborne (2017) explored the likelihood that human jobs will be replaced by
AI technology across 702 occupations in the US labor market. Their findings suggest that
the jobs that are associated with transportation, logistics, administrative support, and
production have a high level of risk to be replaced by AI technology. In hospitality
services, lodging managers, human resources managers, sales and marketing managers,
first-line supervisors, food service managers, and chef and head cooks are considered to be
at low risk (<10%) levels, whereas concierges, customer service representatives, house-
keeping cleaners, and laundry and dry-cleaning workers are riskier, having 21%, 55%,
69%, and 71% probabilities of being replaced by AI devices, respectively. The highest
probabilities (80%+) of replacement are in the line cook, food server, counter attendant
and hostess positions.

Future research agenda

The review of AI literature highlights several research areas that need to be investigated
in the near future in order to fully understand the challenges and impacts of AI as well
as to prepare industry and customers for the changes that will almost certainly come.

AI impact on service encounters

Application of AI to the service industry presents a multi-faceted challenge, as service
appraisals are intertwined with human-to-human interactions and service encounters are
considered to be the most important antecedent in customers’ evaluations of service
performance (Wunderlich, Wangenheim, & Bitner, 2013). Technology acceptance has
been widely studied by researchers, and recent studies suggest that increasing proportions
of consumers are now more willing to accept the use of technology in service delivery
compared to a few years ago, but resistance to change is still significant and can be
a fundamental barrier to IT adoption (Rizzuto, Schwarz, & Schwarz, 2014). Increasing
adoption of AI devices to deliver various services will definitely change how service quality
is viewed and will have significant impacts on customer satisfaction levels – for example, if
automation enables service to be provided at lower costs and with fewer service errors, but
reduces personal interaction and increases technology-related stress, the overall satisfac-
tion level may be higher or lower depending upon the particular customer, specific service
encounter, and price point at which the transaction occurs.

AI acceptance by customers
Limited acceptance of AI and automation have been the main challenge over the past 50 years,
but the limitations have historically centered on utility and capacity. However, as applications
of AI expand and broaden into domains of higher intelligence, the process of acceptance will
become increasingly complicated. Even though a number of theories have been crafted to
776 O. H. CHI ET AL.

explain customers’ acceptance of AI, these theories have not been fully explored, and are based
largely on historical or existing (mechanical) applications of AI with little research conducted
on the potential acceptance of intuitive or empathetic AI applications. Existing theories do not
fully encapsulate the expanded dimensionality that future AI applications will create as
artificial analytical, intuitive, and empathetic intelligences intrude into more personal aspects
of customers’ lives.
A number of potential moderators of the relationship between AI devices and customer
acceptance require exploration, and the list of moderators such as AI device, personal
privacy intrusiveness, or the interaction between device type and intrusiveness will con-
tinue to expand as use of AI applications increase. For example, in the context of
providing an intimate personal service, do anthropomorphic devices create a greater
sense of personal privacy invasion than a disembodied service device? Do the sensors
required to enable voice commands in the bathroom also capture sounds that customers
are not comfortable being stored in a hotel’s databases?

AI acceptance by companies
Researchers have identified the top three barriers to company acceptance of technology as
(1) lack of internal resources, (2) lack of technology, and (3) inaccurate data (West et al.,
2018). However, within the hospitality industry, the differentiation between brand own-
ership and property ownership will add a new fourth dimension. As with any hotel or
restaurant brand initiatives, brand initiatives that are proposed by the people who dictate
how the brand operates and the adoption of those initiatives by owner/investor who make
the financial investments decisions will be determined separately and will be to some
degree a matter of negotiation. Hotel brands will likely recognize the opportunities and
competitive advantages afforded by AI’s service enhancement functions and the role AI
can play in building value in their brands (West et al., 2018), whereas hotel investors are
more likely to be attracted to the cost-saving opportunities and financial benefits that AI
will afford. Costly service-enhancing measures that do not provide financial benefits may
be embraced by the brands but resisted by ownership, whereas cost-saving initiatives that
have strong financial benefits but may detract from customer experiences may be
embraced by ownership but rejected by the brands, highlighting the inherent conflicts
between the two stakeholders. Thus, the path to implementation for technologies that only
pass one of these two tests will require significant research to ascertain how AI decisions
are made within the balance of power between owners and operators.

AI acceptance by employees
In the same manner that resistance to technology by workers hindered the industrial
revolution in the 1800s (Nash & Flesher, 2005), it is possible that employees’ lack of
enthusiasm for AI could undermine the implementation of AI initiatives in the service
industry. Researchers have so far paid little attention to the impact that AI and automation
will have on service employees, and the downstream impacts on service levels moderated
by the attitudes and perceptions of frontline employees. If employees do not accept the use
of AI or if the imposition of AI undermines employee satisfaction or employee–guest
interaction levels, AI could ultimately have a cumulated negative impact on guest
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 777

satisfaction. For this reason, there is an urgent need to investigate hotel employees’
attitudes and acceptance toward AI service devices.
Even in cases where service employees embrace the benefits of artificial intelligence, the
blending of human and nonhuman service agents will require new skills and orientations on
the part of employees. A different skill-set will be required to operate within the higher-tech
environment where routine tasks are handled by AI service devices, and human employees’
roles will become more customer-focused and dynamic, which will require higher levels of
initiative and emotional intelligence. Many line-level service industry employees are not
currently oriented toward “out of the box” thinking and will require retraining.

Balancing human and nonhuman employees

Huang and Rust (2018) proposed that, based on different levels of robotic intelligence, service
robots may empower human employees or serve customers by themselves. To remain
competitive, human employees need to enhance their own skills accordingly. For example,
if robots can handle transactional (e.g. transportation and cleaning) tasks, employees should
improve their analytical skills; if a robot can do analytical tasks, employees need to improve
their empathetic skills or soft skills. However, once the intelligence of service robots reaches
a certain level, which allows these robots to be competent to do empathetic tasks, service
robots may completely replace human employees (Huang & Rust, 2018), leading to massive
unemployment in the hospitality industry. In this case, if AI is adopted without regulations,
the primary benefit may accrue to few people, such as managers and hotel owners
(Nourbakhsh, 2015) while the majority of workers are negatively impacted. Therefore, future
studies should discuss how to regulate the use of AI devices in the hospitality industry and
investigate the potential issues related to the adoption of AI technology.

AI providing social interaction

Even though many existing studies have suggested that AI service devices cannot currently
provide desired social interactions in the service context (e.g., Gursoy et al., 2019; Lin
et al., 2019), Van Doorn et al. (2017)’s conceptual research highlighted that service robots
in the future could constitute automated social presences. Unlike the robots that are
currently available, these automated social presences may enhance rather than undermine
customers’ need for social interaction. In this situation, customers' attitudes toward these
robots is more likely to be influenced by whether they can psychologically accept this new
type of social presence. Further studies are needed to investigate customers’ behaviors
related to AI services in the context that AI devices can provide and enhance social
interactions and explore the conditions and factors that contribute to or detract from
customers’ satisfaction with social interactions with AI devices.

Hotel chain and scale differentiation

Currently, hotel scales and chain affiliations have a significant impact on the rates that
customers are willing to pay, and guest loyalty is rewarded by the hotels by providing
additional amenities (concierge lounge access, upgraded guestroom amenities).
Furthermore, a positive relationship exists between service levels and value, in that
778 O. H. CHI ET AL.

customers are willing to pay premiums for higher levels of personal service. In an
environment where personal services can be provided by devices at little or no cost, the
relationship between service level and price may be altered, with greater (or lower)
premiums associated with human interactions. Guest who seeks luxury experiences may
be more (or less) willing to accept nonhuman service encounters and the value of those
encounters in the minds of consumers may expand or contract.

Transparency, ethics, and fairness

The development of machine learning techniques not only enhances the problem-solving
power of AI devices but also leads to increasing difficulty in understanding the reasoning
process of these devices (Rzepka & Berger, 2018). Increasingly sophisticated AI devices
will generate solutions based upon large datasets and multiple variables combined using
optimization techniques that are not transparent to consumers. A simple example would
be the calculation of driving routes for automated vehicles that include not only data
generated by the vehicle’s sensors but also centralized information on traffic patterns,
police activity, local construction activity, historical fluctuations in driving speeds based
upon weather patterns, etc. along with probabilistic assessments of potential impacts on
driving times from non-routine occurrences such as accidents or changes in the weather –
the optimal route recommended by the AI system may include assumptions about events
that have not occurred and may be counterintuitive to the human consumer. However,
studies rarely explore how AI systems use these variables, leading to a low-perceived
transparency toward these systems (Rzepka & Berger, 2018).
In addition, AI systems “learn” based on programmed datasets or algorithms which may be
incomplete or skewed. For example, an image recognition system used in a hotel may be
trained using a dataset that includes primarily Caucasian customers. The skewed dataset may
enable the system to accurately recognize some customers but not customers of other ethnic
backgrounds (Zou & Schiebinger, 2018). Therefore, the service quality of the hotel may have
unintentional ethnic biases, resulting in critical ethics and fairness issues.

The “dark side” of technology

The collection and storage of vast amounts of personal information has been shown to
have potentially negative consequences which may limit consumers’ acceptance of AI
device use. Past infringements on personal privacy include the multitude of data breeches
reported with regularity in national press, along with specific incidents where companies
have used AI in ways that made consumers feel betrayed are resulted in loss of customer
trust. Examples of the latter include Target using purchase data to predict whether
a woman is pregnant, and insurance companies predicting mortality (Marinova et al.,
2017). More dramatic AI failures have resulted in misdiagnoses of medical conditions,
self-driving vehicles losing control, and death (Yampolskiy & Spellchecker, 2016).
In the hospitality industry, a great deal of information is potentially available regarding
consumers’ behaviors and consumption habits, some of which might be objectionable to
their spouses, doctors, bosses, or communities. Companies implementing new AI features
into their service offerings will need to be mindful of possible intrusion into personal lives
of customers that may not be acceptable. It is critical that thorough studies are conducted
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 779

before the applications are created to establish the extent to which customers are willing to
have their personal information analyzed.

Conclusion
More than 70 years after its creation, artificial intelligence is finally beginning to achieve
its potential and will be transforming society, and with it the service industry, in myriad
ways over the coming decades. One by one, the constraints to implementation of AI and
automation have been overcome, and current levels of technology, data collection, storage,
and processing speeds are more than adequate to support Schwab’s Fourth Industrial
Revolution and usher in profound systemic change and competitive realignments.
Although it remains to be seen how deeply AI will penetrate the service industries,
sufficient studies exist to predict that the impact will be profound and only limited by
customers’ and employees’ acceptance of the role of technology in customer service
encounters. The current study reviewed these studies and identified seven major themes,
which cover the AI research from different perspectives. However, as it is mentioned, the
AI research especially in the service context is still in its infancy stage. Thus, future studies
are urgently needed to further expand the understanding of the impact of AI on different
social groups and the perception of AI technology by different users.
This systematic review also points to the demand for more research in hospitality
literature as the hospitality industry is especially sensitive to the development of AI. As
a high-touch industry, human-to-human interaction is a crucial factor that influences
customers’ perceived service value (Dedeoglu et al., 2018). However, the trend of using AI
devices to perform human tasks and even to replace human employees is likely to
challenge customers’ perceptions of hospitality services and may reform the industry
from high-touch to high-tech. Thus, great effort is needed to investigate the impact of
this technology revolution on the hospitality industry and to explore how to preserve
“touch” when “tech” is being adopted.

References
Ackerman, E. (2016). Study: Nobody wants social robots that look like humans because they
threaten our identity. IEEE Spectrum, 1–5.
Anderson, J., Rainie, L., & Luchsinger, A. (2018, Dec 10). Artificial intelligence and the future of
humans. Pew Research Center. Retrieved from https://www.pewresearch.org/internet/2018/12/
10/artificial-intelligence-and-the-future-of-humans/
Aziz, A. A., Moganan, F. F. M., Ismail, A., & Lokman, A. M. (2015). Autistic children’s Kansei
responses towards humanoid-robot as teaching mediator. Procedia Computer Science, 76,
488–493. doi:10.1016/j.procs.2015.12.322
Bainbridge, W. S., Brent, E., Carley, J., Heise, D., Macy, M., Markovsky, B., & Skvoretz, J. (1994).
Artificial social intelligence. Annual Review of Sociology, 20(1), 407–436. doi:10.1146/annurev.
so.20.080194.002203
Baisch, S., Kolling, T., Schall, A., Rühl, S., Selic, S., Kim, Z., & Knopf, M. (2017). Acceptance of
social robots by elder people: Does psychosocial functioning matter? International Journal of
Social Robotics, 9(2), 293–307. doi:10.1007/s12369-016-0392-5
Bedaf, S., Marti, P., & De Witte, L. (2019). What are the preferred characteristics of a service robot
for the elderly? A multi-country focus group study with older adults and caregivers. Assistive
Technology, 31(3), 147–157. doi:10.1080/10400435.2017.1402390
780 O. H. CHI ET AL.

Belanche, D, Casaló, L.V, Flavián, C, & Schepers, J. (2020). Service robot implementation: a
theoretical framework and research agenda. The Service Industries Journal, 40(3-4),203–225.
Bowen, J., & Morosan, C. (2018). Beware hospitality industry: The robots are coming. Worldwide
Hospitality and Tourism Themes, 10(6), 726–733. doi:10.1108/WHATT-07-2018-0045
Byrne, K. (2018, June 20). Chatbots: Your new best friend? Independent.ie. Retrieved from https://
www.independent.ie/life/chatbots-your-new-best-friend-37023285.html.
Čaić, M., Odekerken-Schröder, G., & Mahr, D. (2018). Service robots: Value co-creation and
co-destruction in elderly care networks. Journal of Service Management, 29(2), 178–205.
doi:10.1108/JOSM-07-2017-0179
Chan, A. P. H., & Tung, V. W. S. (2019). Examining the effects of robotic service on brand
experience: The moderating role of hotel segment. Journal of Travel & Tourism Marketing, 36
(4), 458–468. doi:10.1080/10548408.2019.1568953
Choi, S., Liu, S. Q., & Mattila, A. S. (2019). “How may I help you?” Says a robot: Examining
language styles in the service encounter. International Journal of Hospitality Management, 82,
32–38. doi:10.1016/j.ijhm.2019.03.026
Conti, D., Di Nuovo, S., Buono, S., & Di Nuovo, A. (2017). Robots in education and care of children
with developmental disabilities: A study on acceptance by experienced and future professionals.
International Journal of Social Robotics, 9(1), 51–62. doi:10.1007/s12369-016-0359-6
Crook, J. (2014, August 13) Starwood introduces robotic butlers at Aloft hotel in Cupertino.
Techcrunch.com. Retrieved from https://techcrunch.com/2014/08/13/starwood-introduces-
robotic-butlers-at-aloft-hotel-in-palo-alto/
Daniel, B., Casaló, L. V., Flavián, C., & Schepers, J. (2019). Service robot implementation: A theoretical
framework and research agenda. The Service Industries Journal. doi:10.1080/02642069.2019.1672666
Davenport, T. H., & Kirby, J. (2015). Beyond automation. Harvard Business Review, (June), 59–65.
Dedeoglu, B. B., Bilgihan, A., Ye, B. H., Buonincontri, P., & Okumus, F. (2018). The impact of
servicescape on hedonic value and behavioral intentions: The importance of previous experience.
International Journal of Hospitality Management, 72, 10–20. doi:10.1016/j.ijhm.2017.12.007
Denning, T., Matuszek, C., Koscher, K., Smith, J. R., & Kohno, T. (2009), A spotlight on security and
privacy risks with future household robots: Attacks and lessons. Proceedings of the 11th International
Conference on Ubiquitous Computing, Orlando, Florida, ACM, New York, NY, pp. 105–114.
Edelman, D.C., & Singer, M. (2015, Nov). Competing on customer journeys. Harvard Business
Review. Retrieved from https://hbr.org/2015/11/competing-on-customer-journeys
Fernández-Llamas, C., Conde, M. A., Rodríguez-Lera, F. J., Rodríguez-Sedano, F. J., & García, F.
(2018). May I teach you? Students’ behavior when lectured by robotic vs. human teachers.
Computers in Human Behavior, 80(80), 460–469. doi:10.1016/j.chb.2017.09.028
Fluss, D. (2017). The AI revolution in customer service. Customer Relationship Management,
January, 38.
Foster, M. E, Gaschler, A, & Giuliani, M. (2017). Automatically classifying user engagement for
dynamic multi-party human-robot interaction. International Journal Of Social Robotics, 9(5),
659–674. doi:10.1007/s12369-017-0414-y
Frey, C. B., & Osborne, M. A. (2017). The future of employment: How susceptible are jobs to
computerisation? Technological Forecasting and Social Change, 114, 254–280. doi:10.1016/j.
techfore.2016.08.019
Golden, F. (2014, Nov 18). Royal caribbean's quantum of the seas features a bionic bar with robot
bartenders. Daily News. Retrieved from https://www.nydailynews.com/life-style/royal-caribbean-
introduces-robot-bartenders-article-1.2015449
Gursoy, D., Chi, O. H., Lu, L., & Nunkoo, R. (2019). Consumer’s acceptance of artificially intelligent
(AI) device use in service delivery. International Journal of Information Management, 49,
157–169. doi:10.1016/j.ijinfomgt.2019.03.008
Halzack, S. (2017, January 18). Robots and artificial intelligence set to upend the art of making a
sale. The Washington Post. Retrieved from https://www.washing tonpost.com/news/business/wp/
2017/01/18/robots-and-artificialintelligence-set-to-upend-the-art-of-making-a-sale/
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 781

Hebesberger, D., Koertner, T., Gisinger, C., & Pripfl, J. (2017). A long-term autonomous robot at
a care hospital: A mixed methods study on social acceptance and experiences of staff and older
adults. International Journal of Social Robotics, 9(3), 417–429. doi:10.1007/s12369-016-0391-6
Hennessy, A. (2017, Mar 1). As well or better than humans': Automation set for big promotions in
white-collar job market. CBC News. Retrieved from https://www.cbc.ca/news/business/automa
tion-jobs-canada-computers-white-collar-1.3982466
Hertzfeld, E. (2019, January 31). Japan’s Henn-na Hotel fires half its robot workforce. Hotel
Management. Retrieved from https://www.hotelmanagement.net/tech/japan-s-henn-na-hotel-
fires-half-its-robot-workforce
Holder, C., Khurana, V., Harrison, F., & Jacobs, L. (2016). Robotics and law: Key legal and
regulatory implications of the robotics age (Part 1 of 2). Computer Law & Security Review, 32
(3), 383–402. doi:10.1016/j.clsr.2016.03.001
Hospitality Technology. (2018, November 28). AutoClerk powers new AAHOA property manage-
ment system. Hospitality Technology. Retrieved from https://hospitalitytech.com/autoclerk-
powers-new-aahoa-property-management-system
Hospitality Technology. (2019a, September 11). Pactum launches artificial intelligence tool for
commercial negotiations. Hospitality Technology. Retrieved from https://hospitalitytech.com/pac
tum-launches-artificial-intelligence-tool-commercial-negotiations
Hospitality Technology. (2019b, April 17). Vinpearl hotels and resorts implements facial recogni-
tion tech. Hospitality Technology. Retrieved from https://hospitalitytech.com/vinpearl-hotels-and-
resorts-implements-facial-recognition-tech
Hospitality Technology. (2019c, June 18). BluIP announces RLH corp. is first to use artificial
intelligence virtual agent. Hospitality Technology. Retrieved from https://hospitalitytech.com/
bluip-announces-rlh-corp-first-use-artificial-intelligence-virtual-agent
Hospitality Technology. (2019d, February 04). MSC cruises debuts ZOE, a virtual multi-lingual personal
cruise assistant powered by AI. Hospitality Technology. Retrieved from https://hospitalitytech.com/
msc-cruises-debuts-zoe-virtual-multi-lingual-personal-cruise-assistant-powered-ai
Hospitality Technology. (2019e, July 23). SHR introduces wave RMS. Hospitality Technology.
Retrieved from https://hospitalitytech.com/shr-introduces-wave-rms
Huang, M. H., & Rust, R. T. (2018). Artificial intelligence in service. Journal of Service Research, 21
(2), 155–172. doi:10.1177/1094670517752459
Hudson, J., Orviska, M., & Hunady, J. (2017). People’s attitudes to robots in caring for the elderly.
International Journal of Social Robotics, 9(2), 199–210. doi:10.1007/s12369-016-0384-5
Huet, E. (2016, Oct 19). Pushing the boundaries of ai to talk to the dead. Bloomberg Businessweek.
Retrieved from https://www.bloomberg.com/news/articles/2016-10-20/pushing-the-boundaries-
of-ai-to-talk-to-the-dead
Hyken, S. (2017, June 10). Half of people who encounter artificial intelligence don’t even realize it.
Forbes. Retrieved from https://www.forbes.com/sites/shephyken/2017/06/10/half-of-people-who-
encounter-artificial-intelligence-dont-even-realize-it/#6b6c7163745f
International Federation of Robotics. (2016). Chapter 1.2.1: Definition “Service Robotics” [PDF file].
In Introduction into service robots, pp. 9.. Retrieved from https://ifr.org/img/office/Service_
Robots_2016_Chapter_1_2.pdf
Ivanov, S., Webster, C., & Garenko, A. (2018). Young Russian adults’ attitudes towards the potential
use of robots in hotels. Technology in Society, 55, 24–32. doi:10.1016/j.techsoc.2018.06.004
Javelosa, J. (2017, Mar 30). Major firm announces it’s replacing its employees with A.I. Futurism.
Retrieved from https://futurism.com/major-firm-announces-its-replacing-its-employees-with-a-i
Johnson, H. (2016, May 16). Fast food workers are becoming obsolete. Business Insider. Retrieved
from https://www.businessinsider.com/self-service-kiosks-are-replacing-workers-2016-5
Jörling, M., Böhm, R., & Paluch, S. (2019). Service robots: Drivers of perceived responsibility for
service outcomes. Journal of Service Research, 22(4), 404–420. doi:10.1177/1094670519842334
Kahn, J. (2017, Mar 29). Domino’s will begin using robots to deliver pizzas in europe. Bloomberg
News. Retrieved from https://www.bloomberg.com/news/articles/2017-03-29/domino-s-will-
begin-using-robots-to-deliver-pizzas-in-europe
782 O. H. CHI ET AL.

Kuhner, D., Fiederer, L. D. J., Aldinger, J., Burget, F., Völker, M., Schirrmeister, R. T., …
Burgard, W. (2019). A service assistant combining autonomous robotics, flexible goal formula-
tion, and deep-learning-based brain–computer interfacing. Robotics and Autonomous Systems,
116, 98–113. doi:10.1016/j.robot.2019.02.015
Kunz, W., Kristina, H., Jos, L., & Ben, L. (2018). Future service technologies: Business models,
analytics, and experience. Journal of Services Marketing. Call for Paper, 2017–19.
Kuo, C. M., Chen, L. C., & Tseng, C. Y. (2017). Investigating an innovative service with hospitality
robots. International Journal of Contemporary Hospitality Management, 29(5), 1305–1321.
doi:10.1108/IJCHM-08-2015-0414
Latikka, R., Turja, T., & Oksanen, A. (2019). Self-efficacy and acceptance of robots. Computers in
Human Behavior, 93, 157–163. doi:10.1016/j.chb.2018.12.017
Lee, B., & Cranage, D. A. (2019). Causal attributions and overall blame of self-service technology
(SST) failure: Different from service failures by employee and policy. Journal of Hospitality
Marketing & Management, 27(1), 61–84. doi:10.1080/19368623.2017.1337539
Lee, J. Y., Song, Y. A., Jung, J. Y., Kim, H. J., Kim, B. R., Do, H. K., & Lim, J. Y. (2018a). Nurses’
needs for care robots in integrated nursing care services. Journal of Advanced Nursing, 74(9),
2094–2105. doi:10.1111/jan.13711
Lee, N., Kim, J., Kim, E., & Kwon, O. (2017). The influence of politeness behavior on user
compliance with social robots in a healthcare service setting. International Journal of Social
Robotics, 9(5), 727–743. doi:10.1007/s12369-017-0420-0
Lee, W. H., Lin, C. W., & Shih, K. H. (2018b). A technology acceptance model for the perception of
restaurant service robots for trust, interactivity, and output quality. International Journal of
Mobile Communications, 16(4), 361–376. doi:10.1504/IJMC.2018.092666
Lee, W. J., Kwag, S. I., & Ko, Y. D. (2020). Optimal capacity and operation design of a robot logistics
system for the hotel industry. Tourism Management, 76, 103971. doi:10.1016/j.tourman.2019.103971
Lewis-Kraus, G. (2016, March 2). Check in with the velociraptor at the world’s first robot hotel.
Wired. Retrieved from https://www.wired.com/2016/03/robot-henn-na-hotel-japan/
Li, J. J., Bonn, M. A., & Ye, B. H. (2019). Hotel employee’s artificial intelligence and robotics
awareness and its impact on turnover intention: The moderating roles of perceived organiza-
tional support and competitive psychological climate. Tourism Management, 73, 172–181.
doi:10.1016/j.tourman.2019.02.006
Lin, H., Chi, O. H., & Gursoy, D. (2019). Antecedents of customers’ acceptance of artificially intelligent
robotic device use in hospitality services. Journal of Hospitality Marketing & Management, 1–20.
doi:10.1080/19368623.2020.1685053
Lu, L., Cai, R., & Gursoy, D. (2019). Developing and validating a service robot integration willingness
scale. International Journal of Hospitality Management, 80, 36–51. doi:10.1016/j.ijhm.2019.01.005
Manyika, J., Chui, M., Bughin, J., Dobbs, R., Bisson, P., & Marrs, A. (2013, May). Disruptive
technologies: Advances that will transform life, business, and the global economy. McKinsey
Global Institute. Retrieved from https://www.mckinsey.com/business-functions/mckinsey-digital
/our-insights/disruptive-technologies
Manyika, J., Lund, S., Chui, M., Bughin, J., Woetzel, J., Batra, P., … Sanghvi, S. (2017, November).
Jobs lost, jobs gained: What the future of work will mean for jobs, skills, and wages. McKinsey
Global Institute. Retrieved from https://www.mckinsey.com/featured-insights/future-of-work
/jobs-lost-jobs-gained-what-the-future-of-work-will-mean-for-jobs-skills-and-wages
Marinova, D., de Ruyter, K., Huang, M. H., Meuter, M. L., & Challagalla, G. (2017). Getting smart:
Learning from technology-empowered frontline interactions. Journal of Service Research, 20(1),
29–42. doi:10.1177/1094670516679273
Mcafee, A., & Brynjolfsson, E. (2016). Human work in the robotic future: Policy for the age of
automation. Foreign Affairs, 95(4), 139–150.
Mende, M., Scott, M. L., van Doorn, J., Grewal, D., & Shanks, I. (2019). Service robots rising: How
humanoid robots influence service experiences and elicit compensatory consumer responses.
Journal of Marketing Research, 56(4), 535–556. doi:10.1177/0022243718822827
Mori, M. (1970). Bukimi no tani (the uncanny valley). Energy, 7(4), 33–35.
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 783

Murphy, J., Gretzel, U., & Pesonen, J. (2019). Marketing robot services in hospitality and tourism:
The role of anthropomorphism. Journal of Travel & Tourism Marketing, 36(7), 1–12.
Nadarzynski, T., Miles, O., Cowie, A., & Ridge, D. (2019). Acceptability of artificial intelligence
(AI)-led chatbot services in healthcare: A mixed-methods study. Digital Health, 5,
2055207619871808. doi:10.1177/2055207619871808
Nash, C. Y., & Flesher, D. L. (2005). Employee leasing. The antebellum 1800s and the twenty-first
century: A historical perspective of the contingent labour force. Accounting, Business & Financial
History, 15(1), 63–76. doi:10.1080/09585200500033022
Nourbakhsh, I. (2015). The coming robot dystopia. Foreign Affairs, 23–28. Retrieved from http://
search.proquest.com/docview/1691576666/
Pan, Y., Okada, H., Uchiyama, T., & Suzuki, K. (2015). On the reaction to robot’s speech in a hotel
public space. International Journal of Social Robotics, 7(5), 911–920. doi:10.1007/s12369-015-0320-0
Parmar, S., Meshram, M., Parmar, P., Patel, M., & Desai, P. (2019). Smart hotel using intelligent
chatbot: A review. International Journal of Scientific Research in Computer Science, Engineering
and Information Technology, 823–829. doi:10.32628/IJSRCSEIT
Phillips, J. (2018, Jun 27). Six things to know about Creator, San Francisco’s new burger robot
restaurant. The San Francisco Chronicle. Retrieved from https://www.sfchronicle.com/insides
coop/article/Six-things-to-know-about-Creator-San-13031309.php
Pinillos, R., Marcos, S., Feliz, R., Zalama, E., & Gomez-Garcia-Bermejo, J. (2016). Long-term
assessment of a service robot in a hotel environment. Robotics and Autonomous Systems, 79,
40–57. doi:10.1016/j.robot.2016.01.014
Poggi, J. (2017, Nov). Cmo's guide to chatbots. AdAge. Retrieved from https://adage.com/article/
media/cmo-s-guide-chatbots/307199
Porter, M., & Heppelmann, J. (2014). How smart, connected products are transforming
competition. Harvard Business Review, 92(11), 11–64.
Prentice, C., Dominique Lopes, S., & Wang, X. (2019). Emotional intelligence or artificial intelli-
gence–an employee perspective. Journal of Hospitality Marketing & Management, 1–27.
doi:10.1080/19368623.2019.1647124
PwC. (2019). Hospitality and leisure trends 2019 [PDF file]. Retrieved from https://www.pwc.com/
gx/en/ceo-survey/2019/Theme-assets/reports/hospitality-leisure-trends-report-2019.pdf
Qiu, H., Li, M., Shu, B., & Bai, B. (2019). Enhancing hospitality experience with service robots: The
mediating role of rapport building. Journal of Hospitality Marketing & Management, 1–22.
doi:10.1080/19368623.2019.1645073
Reinders, M. J., Dabholkar, P. A., & Frambach, R. T. (2008). Consequences of forcing consumers to
use technology-based self-service. Journal of Service Research, 11(2), 107–123. doi:10.1177/
1094670508324297
Rizzuto, T. E., & Reeves, J. (2007). A multidisciplinary meta-analysis of human barriers to technology
implementation. Journal of Consulting Psychology, 59(3), 226–240. doi:10.1037/1065-9293.59.3.226
Rizzuto, T. E., Schwarz, A., & Schwarz, C. (2014). Toward a deeper understanding of IT adoption:
A multilevel analysis. Information and Management, 51(4), 479–487. doi:10.1016/j.im.2014.02.005
Robinson, S., Orsingher, C., Alkire, L., De Keyser, A., Giebelhausen, M., Papamichail, K. N., …
Temerak, M. S. (2019). Frontline encounters of the AI kind: An evolved service encounter
framework. Journal of Business Research. doi:10.1016/j.jbusres.2019.08.038
Rodriguez-Lizundia, E., Marcos, S., Zalama, E., Gómez-García-Bermejo, J., & Gordaliza, A. (2015).
A bellboy robot: Study of the effects of robot behaviour on user engagement and comfort.
International Journal of Human-Computer Studies, 82, 83–95. doi:10.1016/j.ijhcs.2015.06.001
Ross, J.R. (2019, Oct 18). 5 ways artificial intelligence offers real roi. Hospitality Technology.
Retrieved from https://hospitalitytech.com/5-ways-artificial-intelligence-offers-real-roi
Russell, S. J., Norvig, P., & Davis, E. (2010). Artificial intelligence: A modern approach. (2nd ed.).
Upper Saddle River, New Jersey: Prentice Hall.
Rzepka, C., & Berger, B. (2018). User interaction with AI-enabled systems: A systematic review of IS
research. International Conference on Information Systems (ICIS), San Francisco, California, Volume:
39.
784 O. H. CHI ET AL.

Schwab, K. (2016, January 14). The fourth industrial revolution: What it means, how to respond.
World Economic Forum. Retrieved from https://www.weforum.org/agenda/2016/01/the-fourth-
industrial-revolution-what-it-means-and-how-to-respond/
Stuart, S. C. (2016, December 5). How do you feel? Affectiva’s AI can tell. PC Magazine. Retrieved
from https://www.pcmag.com/news/how-do-you-feel-affectivas-ai-can-te
Tranfield, D., Denyer, D., & Smart, P. (2003). Towards a methodology for developing
evidence-informed management knowledge by means of systematic review. British Journal of
Management, 14(3), 207–222. doi:10.1111/bjom.2003.14.issue-3
Trejos, N. (2016, Mar 9). Introducing Connie, Hilton's new robot concierge. USA Today. Retrieved
from https://www.usatoday.com/story/travel/roadwarriorvoices/2016/03/09/introducing-connie-
hiltons-new-robot-concierge/81525924
Tung, V. W. S., & Au, N. (2018). Exploring customer experiences with robotics in hospitality.
International Journal of Contemporary Hospitality Management, 30(7), 2680–2697. doi:10.1108/
IJCHM-06-2017-0322
Tung, V. W. S., & Law, R. (2017). The potential for tourism and hospitality experience research in
human-robot interactions. International Journal of Contemporary Hospitality Management, 29
(10), 2498–2513. doi:10.1108/IJCHM-09-2016-0520
Tussyadiah, I., & Miller, G. (2019). Nudged by a robot: Responses to agency and feedback. Annals of
Tourism Research, 78.
Ukpabi, D., Aslam, B., & Karjaluoto, H. (2019). Chatbot adoption in tourism services: A conceptual
exploration. In S. Ivanov & C. Webster (Eds.), Robots, artificial intelligence, and service automa-
tion in travel, tourism and hospitality (pp. 105–121). Emerald Publishing Limited. doi:10.1108/
978-1-78756-687-320191006
Van Doorn, J., Mende, M., Noble, S. M., Hulland, J., Ostrom, A. L., Grewal, D., & Brown, T. (2017).
Domo Arigato Mr. Roboto: Emergence of automated social presence in organizational frontlines
and customers’ service experiences. Journal of Service Research, 20(1), 43–58. doi:10.1177/
1094670516679272
van Pinxteren, M. M. E., Wetzels, R. W. H., Rüger, J., Pluymaekers, M., & Wetzels, M. (2019). Trust
in humanoid robots: Implications for services marketing. Journal of Services Marketing, 33(4),
507–518. doi:10.1108/JSM-01-2018-0045
Victor, A. (2014, November 26). Restaurant employs £6,000 robots to replace waiters and save
money on staff costs … but don’t expect service with a smile. Daily Mail. Retrieved from http://
www.dailymail.co.uk/femail/food/article-2850186/Restaurantemploys-6-000-ROBOTS-replace-
waiters-save-money-staffcosts.html
West, A., Clifford, J., & Atkinson, D. (2018). ‘Alexa, build me a brand’ an investigation into the
impact of artificial intelligence on branding. The Business & Management Review, 9(3), 321–330.
Williams, K. (2018, Dec 27). Biometrics: The “About Face" hospitality needs. Hospitality
Technology. Retrieved from https://hospitalitytech.com/biometrics-about-face-hospitality-needs
Wirtz, J., Patterson, P. G., Kunz, W. H., Gruber, T., Lu, V. N., Paluch, S., & Martins, A. (2018).
Brave new world: Service robots in the frontline. Journal of Service Management, 29(5), 907–931.
doi:10.1108/JOSM-04-2018-0119
Worden, K., Bullough, W. A., & Haywood, J. (2003). Smart technologies. River Edge, N.J.: World
Scientific.
Wu, H. C., & Cheng, C. C. (2018). Relationships between technology attachment, experiential
relationship quality, experiential risk and experiential sharing intentions in a smart hotel. Journal
of Hospitality and Tourism Management, 37, 42–58. doi:10.1016/j.jhtm.2018.09.003
Wunderlich, N. V., Wangenheim, F. V., & Bitner, M. J. (2013). High tech and high touch
a framework for understanding user attitudes and behaviors related to smart interactive
services. Journal of Service Research, 16(1), 3–20. doi:10.1177/1094670512448413
Xu, K. (2019). First encounter with robot Alpha: How individual differences interact with vocal and
kinetic cues in users’ social responses. New Media & Society, 21(11–12), 2522–2547. doi:10.1177/
1461444819851479
 JOURNAL OF HOSPITALITY MARKETING & MANAGEMENT 785

Yampolskiy, R. V., & Spellchecker, M. S. (2016, October 25). Artificial intelligence safety and
cybersecurity: A timeline of AI failures. arXiv.org. Retrieved from http://search.proquest.com/
docview/2080496831/
Yoon, S. N., & Lee, D. (2019). Artificial intelligence and robots in healthcare: What are the success
factors for technology-based service encounters? International Journal of Healthcare
Management, 12(3), 218–225. doi:10.1080/20479700.2018.1498220
Yu, C. E. (2019). Humanlike robots as employees in the hotel industry: Thematic content analysis of
online reviews. Journal of Hospitality Marketing & Management, 1–17.
Zhang, P., & Li, N. (2004). An assessment of human–computer interaction research in management
information systems: Topics and methods. Computers in Human Behavior, 20(2), 125–147.
doi:10.1016/j.chb.2003.10.011
Zhang, P., & Li, N. (2005). The intellectual development of human-computer interaction research:
A critical assessment of the MIS literature (1990-2002). Journal of the Association for Information
Systems, 6(11), 1.
Zou, J., & Schiebinger, L. (2018). AI can be sexist and racist - it’s time to make it fair. Nature, 559
(7714), 324–326. doi:10.1038/d41586-018-05707-8

Appendix A. Numbers of AI-related studies in hospitality SSCI journals.

Journal name Number of studies

Journal of Hospitality Marketing & Management 5
International Journal of Contemporary Hospitality Management 3
International Journal of Hospitality Management 2
Journal of Travel & Tourism Marketing 2
Journal of Hospitality and Tourism Management 1
Annals of Tourism Research 1
Tourism Management 1
Asia Pacific Journal of Tourism Research 0
Cornell Hospitality Quarterly 0
Current Issues in Tourism 0
International Journal of Tourism Research 0
Journal of Destination Marketing & Management 0
Journal of Hospitality & Tourism Research 0
Journal of Sustainable Tourism 0
Journal of Tourism and Cultural Change 0
Journal of Travel Research 0
Journal of Vacation Marketing 0
Scandinavian Journal of Hospitality and Tourism 0
Tourism Economics 0
Tourism Geographies 0
Tourism Management Perspectives 0
Tourist Studies 0
Total 15
786 O. H. CHI ET AL.

Appendix B. Key words used in the paper search.

Artificial intelligence Machine learning Service technology

Robots Human intelligence Technology-based self-service
Technology Automation Smart service
Robotics Singularity Technology adoption
Service delivery Mechanical Intelligence Organizational frontlines
Social Intelligence Analytical intelligence Human–irobot interaction
Social robot Intuitive intelligence Hotel robot
Service robot Empathetic intelligence Adoption
Automation AI devices Smart guestrooms
Services Smart concierge Intention
Consumer behavior Ethics Anthropomorphism
Humanoid