Harnessing Emotional AI to Improve Interactions in Business, Industry and Health Care
Can you imagine a time when your car could sense your mood and take action if you appear to be tired or agitated? Or a time when an online learning experience can detect when you’re disengaged or confused and take steps to reengage or better instruct you? Or a time when technology could be used to monitor people suffering from depression and alert providers or loved ones so interventions could be made?
Rana el Kaliouby can. The YPO member has taken steps to be an integral part of the process of developing this technology. El Kaliouby is co-founder and CEO of Affectiva, a spinoff of MIT Media Lab, which uses AI and advanced technology to evaluate emotional responses among various audiences for marketing and product development applications. She is the author of Girl Decoded. El Kaliouby has been recognized on Fortune’s 40 under 40 list and as one of the Forbes top 50 women in tech. She is a World Economic Forum Young Global Leader and former cohost of PBS’s Nova series on AI. And she’s passionate about the potential applications of emotional AI across a wide range of industries and types of human interactions.
What is emotional AI?
Several years ago Daniel Goleman popularized the concept of Emotional Intelligence (EI), a theory that had originally been forwarded by two psychologists — John Mayer and Peter Salovey. It’s the idea that, in addition to traditional intelligence or IQ, humans benefit and are more successful in their work and personal lives when they also have emotional intelligence — the ability to get along and interact positively with others.
El Kaliouby has extended this concept to the world of AI, theorizing that AI, as it becomes more integral to our lives, will also need to have emotional intelligence, or empathy. “It needs to engage with people just the way we engage with one another,” she says. And, she’s on a mission “to humanize technology before it dehumanizes us.”
Emotional AI and the automotive industry
Consumers have already become quite familiar with interacting with various devices like Apple’s Siri and Amazon’s Alexa. Over the next five years, el Kaliouby predicts, “we’re going to see all our devices interact with us just the way we interact with one another.”
This evolution holds promises for industries, like the transportation industry she says, but also for health care. “Mental health is an area that I’m very, very passionate about and I really think this kind of technology can transform how we measure and how we diagnose mental health,” she says, “even the treatment of mental health.”
Car manufacturers, el Kaliouby says, are adding sensors to vehicles that will be able to detect if drivers are falling asleep or if they’re driving distracted. “We’re able to detect these signs of fatigue, lack of attention, distraction, maybe even frustration or anger,” she says. This can be done, she says, through camera sensors that detect various facial expressions and hand gestures, as well as vocal intonations.
The next generation of vehicles she says, will focus not only on the driver but on the other passengers —so, if children are arguing in the backseat the car could recognize that and attempt to engage them. Or, if someone in the backseat is falling asleep, the car might dim the lights and turn off the music. “There are a lot of features that we’re going to start rolling out that are focused not just on the drive but on the other passengers in the car,” she says.
Emotional AI and the health care industry
El Kaliouby’s work in this area began several years ago while a post-doctoral candidate at MIT focused on the use of automated facial analysis and emotional understanding for children on the autism spectrum who struggle with reading others’ nonverbal cues.
“We piloted a Google glass-like device in 2006 and we deployed it in a number of schools focused on autism and the kids really improved.” Fast-forward to today, she says, and “we are supporting a number of companies that are bringing this to market.” She points to a company called Brain Power as one example.
In another health care application, a company is looking at facial and vocal biomarkers of depression. “That’s an opportunity to kind of learn your baseline and then, if you deviate from that baseline, it can flag that to a doctor or to a family member and you could get support,” el Kaliouby says. She also noted that there are a lot of potential applications in health care for emotional AI technology.
Online education presents another opportunity for this technology, el Kaliouby says. In live classroom settings, the instructor can observe students and interact with them when they feel they may be losing interest or not understanding something. Online, emotional AI technology could do the same thing, she says, perhaps through an online avatar that could detect students’ levels of engagement, confusion or frustration and engage them accordingly.
A nontraditional journey
An Egyptian, el Kaliouby’s journey to becoming a technology expert on the cutting edge of advanced AI and its potential applications across a wide array of industries started in Cairo, studying computer science. Later she moved to the UK for her Ph.D., having been offered a scholarship at Cambridge.
She had just graduated and was newly married when she received the scholarship, but she went to Cambridge on her own with the support of her husband.
It was a lonely experience, el Kaliouby says, and she soon noted that she was “spending more time with my device than I did with any other human being.” That experience, she says, made her realize two things:
- That the computer was completely oblivious to how she was feeling
- That the computer was her main portal of communication with her family back home
But, she says, she felt that “all of the richness of human communication just disappeared in cyberspace.” It got her thinking about how these interactions might be different, and better, if computers could become emotionally intelligent. After spending some time at MIT, she left to start her own company —a move that was a significant risk and outside her comfort zone as she did not come from a family of entrepreneurs and the concept was actually very foreign to her family.
El Kaliouby took a risk, a big one. “I was envisioning this version of the universe that didn’t exist yet and there were a lot of unknowns,” she says. But, she adds, “you just have faith.”
She also ran into obstacles along the way — some technological and some interpersonal and cultural. There were, she says, a lot of people who were naysayers, who said: “You can’t do it. The concept won’t work. You’ll never be able to raise money. You’re a woman.” The list goes on and on, she says. At some point, though, she says, “I just kind of recognized that these obstacles were opportunities.”
But what really drives her, and what she looks for in others, she says, is passion. In fact, passion is a factor she considers when hiring people. “If they’re not passionate about what we’re doing and what we stand for, they’re not a good fit.”
A strong ethical and moral code is also important for any leader but especially when embarking into the unknown area of AI and its potential for both good and ill. El Kaliouby recognizes that dichotomy and the concerns that exist about the ability to quantify human emotional and cognitive states. “There is a lot of potential abuse,” she acknowledges. “We have taken a very strong stance. We talk about the ethnical development and ethical deployment of AI.”
For example, her company doesn’t work in any industry where people can’t give their consent to being recorded or monitored. But even beyond that, she says she believes she has an obligation to “educate the ecosystem of other partners and other technology providers to take the same stance we do.” Her company organizes an annual summit on emotional AI where they share their approach, their philosophy and their strong ethical stance.
Learn more about Rana el Kaliouby’s journey and predictions for the application of emotional AI in this interview with YPO Director of Content, Ekaterina Walter.