Artificial intelligence (AI) has brought about tremendous opportunities for businesses and societies around the world to create greater efficiencies and maximize revenues and profitability.
AI’s potential is still unfathomable as it could expose unprecedented issues and find unexpected solutions faster than any human mind can; this is all due to machine learning (ML).
AI is developing so fast that it is beginning to recognize human emotions by using its own self-taught knowledge. Emotionally intelligent AI could make great strides in improving healthcare systems, marketing campaigns, customer service and the list goes on.
Emotional AI, a growing subset of AI, has the ability to measure, understand, stimulate, react and replicate human emotion. This subset of AI could be used as a tool to enable more natural interactions between humans and machines. This could essentially be viewed as a more human-centric technology whereby it focuses on the emotional response or reaction of the human participant in the equation.
As research scientist at the MIT Media Lab, Javier Hernandez, put it, “Think of the way you interact with other human beings; you look at their faces, you look at their body, and you change your interaction accordingly.”
“How can [a machine] effectively communicate information if it doesn’t know your emotional state, if it doesn’t know how you’re feeling, it doesn’t know how you’re going to respond to specific content?” he added.
It is important to consider, however, that humans are still better at picking up on social cues and reading others’ emotions because human interaction is much more complex than machines can comprehend at this given point in time. Language is much more nuanced that its literal words because meaning is communicated via a variety of things. Meaning is communicated through tone, body language, cultural filters and of course, context.
Teaching machines to pick up on these cues and nuances and comprehend the real message which is being communicated can be quite challenging. However, technology is developing at such an unprecedented rate that we can expect emotional or empathetic AI to improve very soon. Machine learning is making great strides in several processes, so what’s to say that it won’t do so in the case of emotional AI?
Todd Banhidy, founder and chief product architect of Buy It Installed Inc., a company which uses emotional AI in the ecommerce space to enhance customer service engagement, stated, “Artificial empathy is very much like artificial intelligence in that everyone in the industry has their own definition. We use empathy to mean the nonactionable part of being able to tell how a person feels. What you do with that information is the second part. With artificial empathy, a computer can mimic a human’s ability to empathize or share feelings with another. This means a computer can ascertain what a person is feeling with abilities similar to those people use to do the same thing.”
Indeed, if machines learn to present and understand human-like interactions, then they could be incredibly useful tools. Machines are able to collect and analyze large amounts of data. In this case, listening and paying attention to voice inflections could help them better understand what sort of inflections correlate with which emotions. While they are not there yet, machines could soon pick up on subtleties that a person may not even be able to recognize such as micro-expressions on human faces that might happen too quickly for a regular person to recognize.
Empathetic AI has already begun to seep its way into a variety of industries.
What makes AI unique is its consistency, repeatability and scale. It is in fact these features that causes it to begin to understand the role of emotions in customer interactions. Utilizing AI in customer service could completely transform the customer experience.
Self-service channels like chatbots for instance, have gained quite a lot of traction over the past few years, especially in the e-commerce space.
“We have found empathetic AI to be extremely effective in customer service situations. On the phone, a chatbot can pick up on a personality type just through the word choice and tone of voice,” says Banhidy. According to him, the AI asks itself questions such as: “Is the caller someone who likes to chat, or do they just want to get to the point and get things over with? Also, where are they calling from?” and reacts accordingly.
“AI can pick up on these factors and be far more flexible than human customer service reps,” he added.
Scott Sandland, cofounder and CEO of Cyrano.ai, a startup which combines emotional intelligence and artificial intelligence and applies them to commercial environments, echoed Banhidy’s sentiment and stated, “An AI is able to think from a broader perspective than a human salesperson is. Most salespeople are short-sighted. They’re looking for an immediate outcome rather than thinking about the lifetime value of the customer. A skillful AI, on the other hand, can create a sales experience harmonious with the customer’s needs and intentions rather than with the salesperson.”
Consumer research has shown that the usage of chatbots increased by about 10 percent between the end of 2018 and end of 2019. Virtual agents can now hold conversations with us whilst simultaneously performing routine tasks and giving the customer the necessary tools in order to be self-reliant. This not only enables quicker resolutions but also lower costs for companies, leaving call or contact center agents to focus more on escalated or high-impact issues.
Detecting the root cause of customer frustrations can be quite challenging. Outstanding customer service relies on identifying and acting on systemic flaws and vulnerabilities. If the actual source of the customer’s frustration is determined, it would help resolve the issue at hand a great deal faster. Indeed, customer frustration is the most accurate insight into customer satisfaction.
In addition to this, using interaction analytics tools for gaining greater insight into speech and text understanding could unearth some underlying issues such as compliance-related issues and observe the employees better based on the company’s training. Pinpointing these flaws and fixing them could unlock better customer experiences and inherently cultivate a satisfied core audience.
Health and wellbeing
Last year at MWC Shanghai, Telecom review spoke with Rana Gujral, CEO of Emotional AI Company, Behavioral Signals. During the discussion, he speculated the advent of care robots. Using AI to create care robots which could take good care of the elderly. Not only can these robots mitigate loneliness, which is essentially a massive problem now amongst the elderly, but it could also help them with day to day activities such as reminding them to take their medication on time and taking care of their basic needs such as ordering food or re-filling a medical prescription order.
“I recognize that these concepts are still very much in the experimental stage, however, experiences are there and products are available in the market. This can also be said with their challenge to adopt the latest innovations such as smartphones and social media usage, however most of the elderly population have adapted to that,” said Gujral. “There’s always a point where people feel comfortable with the technology and obviously that depends on getting to a basic level of experience in place.”
He added that since we are living in nuclear families where most people might actually struggle to find the time to spend with one another and this is a good resolution to this challenge.
Another use case for empathetic AI in the wellbeing space could be for therapy. An AI-powered robot therapist, at first glance, seems particularly dystopian but if emotional intelligence becomes the most prominent part of the equation, it may well be the way forward.
Using robots for therapy would essentially remove the bias that individuals often carry with them.
As Sandland put it, “Let’s imagine a therapist that has the best training and remembers everything and always cares. Now, this is a best-case scenario that never happens, but let’s say it does. Even then, a human ego can contaminate patient outcomes. Over time, a human’s opinions can ossify. They can become close-minded. There will be times when even the best therapist is tired or hungry or eager to leave on vacation. Sometimes, the human therapist is just going through a bad day.”
As of yet though, humans are the best option when it comes to therapy but with the advancement of human-centric AI, this is no longer a far-off dream.
From data privacy to ensuring decisions are made by humans, emotionally intelligent AI could bring about some concerns.
In contrast to this, empathetic AI is a much more valuable asset to have as opposed to AI that is not emotionally-savvy.
In reference to this, Gujral stated: “If you have a human whom you depend on and who’s very intelligent, but is not emotionally aware, that is the clinical version of a psychopath. Would we want machine psychopaths? An emotionally intelligent machine would typically be more ethical and fairer rather than a very intelligent machine you depend on but who has no ability to process emotions.”