Toward Empathetic Tech: AIHce Keynoter’s Optimistic Vision of Our Digital Future
By Ed Rutkowski
In the months since ChatGPT took the internet by storm, much has been written about the perils of artificial intelligence, with many observers warning that humans have ceded too much authority to AI. Poppy Crum isn’t worried about that. If anything, as Crum explained during yesterday’s opening keynote address at AIHce EXP 2023 in Phoenix, she wants AI to be smarter and more powerful, and she foresees a near future where our environments adapt to us instead of the other way around.
Such a future is possible because of advances in sensor technology that provide ever-deeper insights into the varieties of human experience. “We are getting to the point of really understanding an individual in a dynamic way that we can use to really improve the quality of life,” Crum told her audience at the Phoenix Convention Center and hundreds of virtual attendees.
The key to this transformation, Crum said, is the ongoing development of “empathetic technology,” a term for devices that respond to biological measurements. Examples of such devices can be intelligent thermostats that sense whether the occupants of a space feel warm or cold and modulate the temperature accordingly. In a workplace setting, sensors that detect the dilation of pupils—a sign of high cognitive load—could alert OEHS professionals that an employee is under significant stress.
There are many such biomarkers that can be put to useful purposes. A tiny sensor placed in the inner ear, for example, can create an electroencephalogram that potentially identifies the biomarkers for seizures or strokes. Crum said that, of the approximately 1 million strokes diagnosed per year in the United States, about one third have no symptoms. Because an individual who experiences a small stroke is highly likely to have a large, devastating stroke in the future, technology that detects the precursor events can alert people to the need for treatment, possibly saving lives.
Technology could also be deployed to act on recent advancements in our understanding of how human speech reflects health. Devices that analyze speech patterns could predict the development of Alzheimer’s by picking up on verbal cues that escape human recognition, such as slight stutters, repetitions, and delays between sentences. Other conditions, including multiple sclerosis, diabetes, and heart disease, also have warning signs that show up in a person’s voice, Crum said.
Even the ways people interact with their smartphones can yield insights into health. What Crum referred to as “swipe dynamics”—the patterns a finger makes as it moves across a screen—have been shown to predict a person’s cognitive state and could be used to identify signs of bipolar disorder.
If the benefits of such applications are obvious, so are the hurdles, particularly ethical considerations. Crum acknowledged that people may not want technology to analyze their every word or map their every biological signal. Figuring out how to give society agency over these matters will be crucial to success; Crum suggested that digital reconnaissance be kept out of the workplace, for example. But she implied that concerns about privacy will resolve as people become more familiar with the advantages of empathetic technology.
A similar transformation has already occurred in attitudes toward the internet. As Crum noted, parents in the early digital age commonly warned children not to talk to strangers online or to get into strangers’ cars. “And now,” Crum said, “we literally summon strangers on the internet and get into their cars.” The taboos couldn’t withstand the convenience of services such as Lyft and Uber.
“The developments in sensors that allow us to know things about our bodies in our spaces are exponential,” Crum said. It seems only a matter of time before their potential, and their consequences, become reality.
Ed Rutkowski is editor in chief of The Synergist.