Skip to main content

Your A.I. smart assistant could one day tell if you’re lonely

Your A.I. assistant can do plenty of things for you, whether that’s answering questions, cuing up the perfect song at the right time, making restaurant bookings, or many other tasks. Could it also work out whether you’re lonely?

“Emotion-sniffing” technology is a growing area of interest among researchers, but it’s still in its infancy. In a new proof-of-concept study, researchers from the University of California San Diego School of Medicine recently showed how speech-analyzing A.I. tools can be used for predicting loneliness in older adults.

“Our study of loneliness in older adults applied A.I. technologies called Natural Language Processing (NLP) that look for subtle linguistic patterns such as expressed sentiment and emotion,” Ellen Lee, assistant professor of psychiatry, told Digital Trends. “We analyzed data from 80 interviews conducted with older adults, ranging in age from 66 to 94 years, and examined how language features predict either self-acknowledged loneliness or loneliness based on a validated scale.”

The interviews used as the basis of the study were conversational in nature, and lasted up to 90 minutes. They were accompanied by a more conventional loneliness assessment, answered by the participants. Interviews were then transcribed and analyzed using an NLP tool developed by researchers at IBM. The A.I. algorithm the researchers developed was able to predict self-acknowledged loneliness with an accuracy level of 94 percent, and loneliness based on the loneliness assessment scale with a precision level of 76 percent.

Interestingly, the study shed some light on differences in the way that different people, such as men and women, acknowledge loneliness and express emotions about it. Were a tool such as this to ever roll out into a commercial product, or at least one that was publicly available, it would be crucial to acknowledge these differences.

Engaging users in daily discussions

“A voice-activated home assistant like Alexa or smartphone assistant like Siri could [one day be used to] engage older adults in daily discussions that would help to establish a baseline of expressed emotions and speech patterns, as well as be able to flag changes that might indicate increasing loneliness,” Varsha Badal, a postdoctoral research fellow who worked on the project, told Digital Trends. “Ideally, interventions for loneliness could be delivered remotely, allowing us to help older adults who are not coming into the clinics or are staying home more than usual due to the COVID-19 pandemic or other concerns.”

Next up, the researchers will be combining their algorithms with other types of assessment, such as sleep and mobility data, to further develop the project. They will also test it on larger populations to ensure its accuracy.

This is still early days for the research. There are plenty of ethical issues when it comes to, for instance, using robots and A.I. to care for the elderly. (Would an A.I. assistant use its loneliness-sensing ability to somehow initiate more social contact or just to fire up a jokes playlist?) It’s also not clear how users would feel about their A.I. assistants making judgments about their mood from a privacy perspective. Nonetheless, it’s fascinating research — and an area we’ll likely see a whole lot more of in the years to come.

A paper describing the work was recently published in the American Journal of Geriatric Psychiatry.

Luke Dormehl
I'm a UK-based tech writer covering Cool Tech at Digital Trends. I've also written for Fast Company, Wired, the Guardian…
This clever new A.I. assistant will screen and block robocallers for you
fcc robocall blocking legistation robocalls mem 2

Robocalls are a massive problem, with the average American citizen being on the receiving end of an average of 18 of these automated calls every month. While block lists can put a stop to some of those, they remain a big headache for most people. What if you could turn to yet more robots, these ones fighting on the side of good, to help?

That’s the idea behind a new project carried out by researchers at the Georgia Institute of Technology. They have created a prototype virtual assistant -- think Google’s Duplex or Amazon’s Alexa -- that could act as intermediary between a caller and the recipient. In essence, it’s a bit like a robo version of a secretary, who answers the phone for you, and then passes it over if the call turns out to be worth taking. In tests involving some 8,000 robocalls, the virtual assistant was able to block every single one. When it was put through its paces with human participants, it was 97.8% effective at recognizing them as non-bots.

Read more
A.I. can tell if you’re a good surgeon just by scanning your brain
brain with computer text scrolling artificial intelligence

Could a brain scan be the best way to tell a top-notch surgeon? Well, kind of. Researchers at Rensselaer Polytechnic Institute and the University at Buffalo have developed Brain-NET, a deep learning A.I. tool that can accurately predict a surgeon’s certification scores based on their neuroimaging data.

This certification score, known as the Fundamentals of Laparoscopic Surgery program (FLS), is currently calculated manually using a formula that is extremely time and labor-consuming. The idea behind it is to give an objective assessment of surgical skills, thereby demonstrating effective training.

Read more
Facebook A.I. could fix one of the most annoying problems in video chat apps
Woman looking at videos on Facebook

Communication on Facebook might be predominantly carried out via text, but the social media giant may nonetheless help to solve some of the biggest challenges with audio communication. Announced on Friday, July 10, ahead of the International Conference on Machine Learning, Facebook has developed a new, cutting-edge artificial intelligence that’s able to distinguish up to five voices speaking simultaneously.

That could be transformative for everything from next-gen hearing aids or smart speakers dialing in and amplifying certain voices to future Zoom-style video conferencing learning to better prioritize speakers to stop everyone talking over each other.

Read more