Privacy Breaches: Google Could Access Entire Chat History

Google announces AI as a new feature to their chat product which can access your entire chat history for “learning” purposes.

February 5, 2024

By: Bobby Casey, Managing Director GWP

google I’m all for a private party or business calling their own shots. Your property, your rules. It’s simple, and it’s the right thing to do. I defend that because I exercise that same discretion. I might not have the same rules as others, but that is as much my right as if I had more rules than they do.

And in a free market situation, this wouldn’t happen unquestioningly.

But in the current fascistic circumstances, what we have is private industry carrying out government whims.

  • The government demanded private retail companies enforce distancing and masks during the pandemic.
  • You have the Twitter files where private social media companies were censoring stories at the behest of the FBI.
  • You have private biometric companies selling their services to law enforcement.
  • You have private tech companies building in government mandated “back doors” to people’s data.

It’s time to stop pretending large “private” industries are anything other than government stooges.

Artificial Intelligence is getting a lot of play and attention right now. It’s hot. It’s new. And in some ways, it’s proven to be a very useful tool. The fuel of excitement tends to push these sorts of novel things forward faster than the throttle of caution or concern can slow things down.

Tech giants still haven’t solved for all the swirling privacy concerns and breaches in their earlier breakthroughs, and already we’re seeing these headlines:

Google update reveals AI will start reading all your private messages

This isn’t just about browser searches with some algorithmic learning. This is machine learning tone, context, and nuance in a chat exchange:

“…[I]t may ask to analyze your messages “to understand the context of your conversations, your tone, and your interests.” It may analyze the sentiment of messages, “to tailor its responses to your mood and vibe.” And it may “analyze your message history with different contacts to understand your relationship dynamics… to personalize responses based on who you’re talking to.”

It might be convenient to have AI handle your communications from emails, to texts, to even resume building, but AI can only build on what it knows. It has to get its instruction from somewhere.

How will all this get processed then? Google is very cloud based, so it comes as no surprise that’s where your data will go first:

These will be sent to the cloud for processing, used for training and maybe seen by humans—albeit anonymized. This data will be stored for 18-months, and will persist for a few days even if you disable the AI, albeit manual deletion is available.

Such requests fall outside Google Messages newly default end-to-end encryption—you’re literally messaging Google itself.

It was only a few years ago (2020) when Amazon, Apple, and Google were outed for eavesdropping on conversations being had in private settings because the device mistakenly thought it was being alerted to start recording.

“[T]hey found the devices were activating without the wake word being uttered between 1.5 and a whopping 19 times a day. That’s potentially a lot of recordings of conversations that might be very private. They found the HomePod and Cortana devices activated the most, followed by Echo Dot series 2, Google Home Mini, and Echo Dot series 3.

Worse, the researchers discovered that the devices can be recording for quite a long time, allowing them to pick up conversations.

  • Echo Dot 2nd Generation and Invoke devices had the longest activations of 20 to 43 seconds. HomePod and the majority of Echo devices saw activations lasting six seconds or more.
  • The Google Home Mini was often mistakenly activated when words were spoken that rhymed with “hey” such as “they”, followed by something beginning with the letter “G” or something that contains “ol” such as “cold.” The researchers discovered that “I can spare” and “I don’t like the cold” both set off Google’s device.
  • Apple’s HomePod was often activated when words rhymed with “hi” or “hey” followed by something starting with “S”, or when the syllable rhymed with “ri.” For example “and seriously”, “I see”, “I’m sorry”, “they say.”
  • Amazon devices were activated when words contained a “K” sound and were similar to “Alexa”–for example “exclamation”, “Kevin’s car”, “congresswoman.”
  • Microsoft’s Invoke was activated with words starting with “co”, such as “Colorado”, “consider”, “coming up.”

It wasn’t that long ago when Ring was caught spying on users in their homes, whereby employees and third-party contractors were given unfettered access to video footage of Ring users. It was only last year that this finally was reeled in.

Never forget the man who was locked out of his Echo account because a black Amazon delivery man insisted he was yelling racial slurs at him. Only to find out the man in question was himself also black, and NOT home at that time. That took two weeks to sort out.

It’s not just that the company can see your information, it’s what can be done with it once they have access to it.  Can they sell it?  Can they threaten you?  Can they openly share it with the government?

Zak Doffman offers a very prescient and wise word of caution toward the end of his article: There’s a difference between what can’t be done, such as breaching end-to-end encryption, and what isn’t being done, such as policies as to where content analysis takes place.

Like anything, there are noble and ignoble applications for AI. If the capability exists to do the noble thing, that’s great. But if there are no safeguards to prevent the ignoble thing, that’s bad.

The more cynical take is that if they are just now telling you about this, chances are they were doing this for much longer. Can I prove that? No. But given all the other breaches in user privacy, it’s not entirely implausible. Regardless of how long this has been happening or if it has yet to happen, the important part is to keep in mind is the capability. Just because it isn’t being done, doesn’t mean it can’t be done.

The need to be vigilant is ever-present.

Click here to get a copy of our offshore banking report, or here to become a member of our Insider program, where you are eligible for free consultations, deep discounts on corporate and trust services, plus a host of information about internationalizing your business, wealth and life.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top



Privacy Policy: We hate SPAM and promise to keep your email address safe.


Enter your name and email to get immediate access to my 7-part video series where I explain all the benefits of having your own Global IRA… and this information is ABSOLUTELY FREE!