Perhaps no matter what companies say about protecting user privacy, no one can give a guarantee that that a consumer’s privacy can’t be breached. The same happened even with the Tech- Giant Google, which had just recently launched its next generation smart speakers, dubbed Google Home Mini, the past week.
Even before the expected launch date of 19 October, the voice enabled speaker Home Mini was found to have a bug in its software. This particular bug was discovered by a renowned journalist Artem Russakovskii who works at Android Police.
Launched at an event at San Francisco for a price tag of a mere $49, the Google Home Mini was provided to the members of the press home team as a review unit, and Artem was one of the members.
Unlike other the standard home assistants, which operate on demand, by just calling out a specific phrase like the “Hey Google” for Google Homes, the one Artem got that day and placed in his bathroom was quite different. The usual home assistant, just switches itself on, waits for us to ask a query, and then responds back by getting the specific order from the cloud.
On the contrary, Artem’s Home Mini was able to turn itself on without being ordered to do so, and was able to quietly record any sounds which may be coming in its direction. Not only did it just record the sounds, but was constantly sending it to the Google’s servers, since basically that’s what the job of a home assistant is going to be, but the problem which remained was that in this way every possible voice, could be sent to the cloud, arising privacy concerns.
According to Artem on his website, “My Google Home Mini was inadvertently spying on me 24/7 due to a hardware flaw,”. The curious thing it was able to do so, without creating any sound, or the casual reply back. The only thing which gave it away were the four lights on the Mini, which indicated that the Mini was being constantly switched on.
Then further to confirm his suspicion, Artem decided to check his Google account for ‘My activity’ which enlisted all the times when the device had been switched on inadvertently without even being called upon to activate by the unique hotword.
Therefore, the only option left for him was to get in touch in Google and ask for their assistance. The Google team was responsive and was helpful in fixing the problem. According to a Google spokesperson, “We learned of an issue impacting a small number of Google Home Mini devices that could cause the touch mechanism to behave incorrectly. We rolled out an update on October 7 to mitigate the issue, before Google Home Mini shipped to anyone in the UK,”.
To put it in simple words, this issue was prevalent only in the review units distributed at the U.S Made by Google launch. Every standard home assistant, has inbuilt support such as a button to activate the respective assistant without calling out the hotword for when you don’t want to make noise. As far as the Google Home mini is concerned the same applies to this smart assistant in which case it is invoked by long pressing on the touch panel.
The issue was that Artem’s and other review units had a glitch with the touchpad rendering their devices to be turned on continuously. The issue seems to have a temporary fix ; just disable the touch feature. The issue didn’t resolve here. People were concerned about their privacy and hence Google had to delete any activities created by the inadvertent long presses between the 4th and 7th of October.
If you still are concerned about your privacy you can always deactivate any existing Google devices from capturing sounds, via its settings page.