My father wants to grow old in his place. Artificial intelligence will be watching

Weeks later, out of curiosity, I asked for copies of everything Sensei had been recording at my parents’ house. As I read his personal conversations, I suddenly felt like a spy, the machine my silent co-conspirator. I had lobbied for this in the first place, but now I felt uncomfortable about it. Meanwhile, my father did not remember being told that Sensei was eavesdropping on his conversations.

When I read his words to him, I prepared for the worst.

“So, what do you think?” I asked.

There was a moment of silence when I heard the blood swirling in my ears.

“Okay,” he said finally, his voice strained. “It’s so weird that he hears the words.” He seemed puzzled that anyone would consider his conversations worth copying.

“But I think it’s worth it,” he added before changing the subject.

This story is part of The future of the housea collaboration between WIRED and Architectural Digest editors to help you understand what “home” will look like tomorrow and beyond.

After my father Ignoring the admission, I started looking for what she had already put in his house. I learned that Sensi is one of a growing number of AI devices targeting seniors: Erez and Ally cares Monitor nursing home residents for coughing, falls, and atypical movements, while… Cherish the calm— which looks like a sleek, old-fashioned home speaker — uses radar to detect when someone in the room has fallen or fallen. (It could be the device Bundled with AT&T For rapid response to emergency situations.)

And unlike Alexa, these devices don’t wait for someone to say “help.” Instead, they start recording after specific events: sounds like blows, coughs, screams, and movements like falling out of bed. In Sensei’s case, the device doesn’t tell the senior clerk that he’s recording, which helps explain my father’s confusion.

Claimed to be based on “1,000 years” of audio data, Sensei’s algorithm claims to identify deviations in a person’s usual routines. If you have a new cough, are always in the bathroom, or are scurrying around the house in a new way, Sensei can clearly tell. But when I asked Romy Jobs, the company’s co-founder and CEO, how the algorithm was built, she said only that its models are “trained on anonymized datasets” stripped of “personally identifiable information.” It did not explain exactly what these data sets contained or where they were pulled from.

Sometimes the machine works exactly as it should, Steve Kamau, the calm, soft-spoken coordinator at Husky Senior Care, the agency that helps my father with shopping and other household tasks, told me. In one case, an elderly person fell while trying to reach the toilet when no caregiver was present. Sensei picked up the sound of the crash and the man’s cries for help. Kamau called the agent (who always kept his phone on him) and confirmed his fall, then dispatched 911; The man was then helped off the ground. In another case, he says, a client’s cough was detected early enough that it may have saved her from a more serious illness. (Sensi claims a 90 percent accuracy rate, with end states reviewed by a “human in the loop”; Kamau told me the system also mistook the fallen remote for a fallen senior.)