Future 2020: Will robots replace humans?

The most pressing question of the decade: Will robots replace humans? Are we heading towards Robonomics?

We have all grown up watching movies such as The Terminator, Star Wars, Chappie, RoboCop amongst many others, that have rooted in us the fear of being replaced by robots.  This fear of being the subservient one has considerably increased the research into this field. Every day there is at least one article on how technology is changing the way we work and the way we live, usually accompanied by a list of occupations that will be replaced by automation and robots, and the skills that will be irreplaceable in the world dominated by machines.

Amongst industrial robots, professional service robots, and personal robots, what sparked my interest was personal robots, specifically whether robots can replace humans at the concierge desk. Even though a personal robot has the moderate capability of social interaction, mobility, and is somewhat autonomous, is it really capable of replacing a human concierge, who can personalise their response according to a guest’s requirements and offer them empathy and comfort in complex environments?

According to the Technology Acceptance Model, an individual’s information systems acceptance is determined by two major variables:  Perceived Usefulness (PU) and  Perceived Ease of Use (PEOU), with perceived ease of use having a direct influence on perceived usefulness. Apart from perceived usefulness and ease of use, humanlike appearance (Anthopormorphism) is a major cause of positive perceptions and attitudes among consumers. In saying that, the “Uncanny Valley Theory” says if the machines appear quite close to the human appearance, humans find it eerie. Therefore, robots need to be designed very carefully in terms of their appearance. Apart from the appearance, the perceived intelligence of the robot and perceived safety are also said to induce the acceptance of robots.

The robot that I will be discussing here is a Social Robot, a type of personal robot. Social robots have the capability of autonomous decision making based on the data they receive by various sensors and other sources such as Pepper and Nao. In a customer service setting, a robot assumes a front line role as an interaction counterpart of the customer, making the consumer feel like they are in the presence of another social entity (van Doorn et al., 2017).

Let’s look at three examples to understand how social robots have been employed in the workplace: 

Henn-na hotel: The Henn-na Hotel in Japan employed robots to become “the most efficient hotel in the world” by reducing manpower and having 90% of staff be robotic. By employing the robots at the reception, as a concierge, as robot porters, and in the rooms, Henn-na hotel became the world’s first hotel staffed by robots, entering the Guinness book for world records. However, the success of the hotel was short-lived. The concierge robot found it difficult to answer questions, the dancer robot broke down, and the question and answer robot found it difficult to understand the guest’s questions. Hence, half of the robots had to be laid off as it increased the work of humans rather than reducing it. According to Ms. Matsuo from Henn-na hotel, “it is important to separate the services performed by robots and the services performed by humans.”

Since a robot can’t guess what a guest wants and be empathetic and provide comfort to guests in the same way a human can, it’s crucial to delegate work based on the robot’s capabilities. 

Connie at Hilton: IBM and Hilton created Connie in 2016 and tested the artificial intelligence-powered concierge robot for its chain of hotels in Hilton McLean in Virginia. Connie which is based on Nao, a humanoid robot, answered questions about hotel information, directions, and tourist attractions. Its a 2 feet tall bipedal robot with a human form that has two eyes, a head, arms, and legs. Since its an AI robot, it can learn from its previous experience and personalise the answers for the guests.

Pepper at Westfield:  Five Pepper robots were deployed in Westfield centers in San Francisco. It delivered valuable insights about the shopping center and took some load off the concierge. The idea was to keep the customer journey memorable and keep them coming for more. The tablet on Pepper’s chest showed a menu of options, it was programmed to say hello in six languages, and had a customer service survey. Pepper was designed with a familiar resemblance to Nao but the human likeness was avoided to avoid falling into the “Uncanny Valley Theory.” Pepper is gender-neutral with big manga-like eyes influenced by Japanese culture and a hip joint so Pepper can bow down upon meeting someone.

So, why were these robots employed to replace humans? What are the benefits and costs of employing robots apart from it being a novelty?

The benefits can be divided into financial and non-financial benefits. Financial benefits include labor cost savings as robots can operate 24/7 without getting tired. Since robots do not have to deal with emotion, they perform thousands of tasks without complaints and without forgetting to do it. Non-financial benefits include improved brand image through the adoption of innovative ways of serving customers, communicating and engaging with customers. Using service/social robots definitely creates a unique experience for the guests. A robot can communicate in as many languages as you program it to whereas it’s not always possible for their human counterparts to do that. It does not just save time from performing tedious and repetitive tasks but also adds a fun element in the service delivery process.

If there are benefits, there are also costs involved in deploying robots. From the initial acquisition costs and installation costs, it comes with maintenance costs. As in the case of the Henn-na hotel, it was easier for the hotel to lay off the robots rather than replacing them due to high costs. Even though robots have partial mobility, the premises need to designed to facilitate the robot’s mobility. Both Nao and Pepper robot are equipped to handle basic tasks but they can’t handle ambiguity. Hence, rises a need to train staff and hiring specialists.

Considering the costs and benefits of a robot operated workplace, it is yet to be established whether robots can truly become advanced enough to replace human concierge. The robot revolution is in its infancy and these experiments are laying the foundations for Robonomics.

Robots are not ready to replace humans yet but how much human can they become? I guess we will find it out in the next 10 years. Until then we will continue our experiments!

———————

I will be a candidate of Master of Research at Western Sydney University from February 2020. The focus of my research is to establish whether robots can replace humans at the concierge desk. If you would like to collaborate with me on this project, please email me or send me a LinkedIn message.