As many of us settle into the new year with virtual assistants like the Amazon Echo, Google Home or Microsoft Cortana, it’s worth asking: how much personal information are they picking up? Also, where does that information go?

Perhaps you’ve seen the ads, like this one for Google Home.

Consumer Watchdog, a Santa Monica-based nonprofit, has asked federal regulators to get out in front of a future where virtual assistants are listening to everything that goes on in your home, whether or not you use the “wake” words to launch them.

John Simpson of Consumer Watchdog has read the patent applications for these assistants, and he says those applications clearly indicate plans for broader, commercially focused household surveillance in the near future. “You know, keeping track of things like the number of times you flush the toilet, and when you go to bed,” he says.

Simpson adds that kids using your devices are being tracked, too, regardless of whether you’re in the room or giving permission. He asserts that such tracking is a violation of the Child Online Privacy Protection Act. “One of the patents proposes inferring whether a child is misbehaving,” Simpson says. “I just think that has crossed a line and is far too invasive.”

The Federal Trade Commission would only confirm that it had received a letter from Consumer Watchdog.

Companies Respond

In response to Consumer Watchdog, Google issued a statement. “Consumer Watchdog’s claims are unfounded,” it reads. “All devices that come with the Google Assistant, including Google Home, are designed with user privacy in mind. For Google Home, we only store voice queries after a physical trigger or after recognizing a hot word trigger like ‘OK Google’ or ‘Hey Google.'”

The answer from Amazon spokesman Leigh Nakanishi was more open-ended. “We take privacy seriously and have built multiple layers of privacy into Echo devices,” she wrote. “Like many companies, we file a number of forward-looking patent applications that explore the full possibilities of new technology. Patents take multiple years to receive and do not necessarily reflect current developments to products and services.”

Microsoft declined to comment.

Virtual assistants on display at the Amazon store in San Jose.
Virtual assistants on display at the Amazon store in San Jose. (Photo: Rachael Myrow/KQED)

All of that said, Big Tech is already tracking you: your searches, purchases online and off, social media browsing, even email. You want out of this data-mining ecosystem? It’s going to take work.

As Brian Blau, a VP at the tech research firm Gartner explains, “You’d have to leave your smartphone away from where you are, drive a car anonymously, or not your own car, drive in someone else’s. Not walk in front of any cameras anywhere. You know, there are millions and millions of them all over the place.”

Blau has several virtual assistants in his home, for research purposes. “I have to be careful, because if I say their names right now, they will respond accordingly,” he jokes.

But he’s not worried about them tracking his every move and sending the information to advertisers. What worries Blau is the likelihood that, advertisers will eventually want to talk to you directly through your assistant.

For those creeped out by the thought of recorded requests and other things you’ve said being stored in a database, you can mute your device’s microphone, turn off the device when you’re not using it, or periodically delete the files being kept on you.

But as Amazon warns, “deleting voice recordings may degrade your Alexa experience.”

How Paranoid Should You Be About Your Nosy Virtual Assistant? 4 January,2018Rachael Myrow

  • It appears to me that recording voices unbeknownst to the speaker is a clear violation of California Penal Code § 632. To quote the Digital Media Law Project, “California’s wiretapping law is a ‘two-party consent” law. California makes it a crime to record or eavesdrop on any confidential communication, including a private conversation or telephone call, without the consent of all parties to the conversation.” The issue is that even though I may have consented deep in the bowels of some obscure online disclaimer to being recorded, others that get recorded by, e.g., my phone have NOT so consented. For example, after forgetting to turn my phone off during a recent doctor visit, I found that my phone had recorded and was doing searches on terms my doctor had used in my visit.

    I consider that really scary. Clearly, my doctor had not consented to being recorded nor could I consent for her. Further, it became obvious that Google was trading on (and making money from) what I consider to be illegally collected and private medical information. I’m not a lawyer, but I would think that was a violation of HIPAA.

    I would think this might lead to a massive class action suit (or two or three).


Rachael Myrow

Rachael Myrow is KQED’s Silicon Valley Arts Reporter, covering arts, culture and technology in San Mateo, Santa Clara and Santa Cruz Counties. She regularly files stories for NPR and the KQED podcast Bay Curious, and guest hosts KQED’s Forum.

Her passion for public radio was born as an undergrad at the University of California at Berkeley, writing movie reviews for KALX-FM. After finishing one degree in English, she got another in journalism, landed a job at Marketplace in Los Angeles, and another at KPCC, before returning to the Bay Area to work at KQED.

She spent more than seven years hosting The California Report, and over the years has won a Peabody and three Edward R. Murrow Awards (one for covering the MTA Strike, her first assignment as a full-time reporter in 2000 as well as numerous other honors including from the Society of Professional Journalists, the Radio Television News Directors Association and the LA Press Club.
Follow @rachaelmyrow

Sponsored by

Become a KQED sponsor