I Downloaded My Google Data. Here’s What 12 Years of My Life Looks Like to a Machine

I Downloaded My Google Data. Here’s What 12 Years of My Life Looks Like to a Machine.

By Akshay A. Walimbe

Last month, I did something that I recommend you do too. Something that will take you about thirty seconds to initiate and about two days to fully absorb.

I downloaded my Google data.

If you have a Google account, you can do this right now. Go to takeout.google.com, select “Export all,” and wait. Google will package everything it has on you into a neat little archive. When my archive was ready, I clicked download.

The file was 147 gigabytes.

Let me say that again. One hundred and forty seven gigabytes. Of me. My searches, my emails, my photos, my locations, my YouTube history, my voice recordings, my purchases, my calendar, my contacts, my Chrome browsing history, my Maps timeline, my Drive documents, my Fit health data. Twelve years of my life, compressed into a file larger than most hard drives could hold a decade ago.

I sat there staring at the download progress bar, and a thought occurred to me that I had never had before: I do not remember most of what is in this file. But Google does. Every single bit of it.

The Categories

Let me walk you through what Google has, not the specific content of my data, but the categories. Because the categories alone are enough to make you reconsider everything you think about digital privacy.

Search History. Every question I have ever asked Google. Medical symptoms I looked up at 2 AM. Restaurants I searched before a date. Salary benchmarks I checked before a negotiation. That embarrassing thing I searched out of curiosity and immediately regretted. All of it. Going back years. Not just the search terms, but the time, the device I used, and what I clicked on afterwards.

Location History. A complete map of everywhere I have been, timestamped down to the minute. My commute to work. The hospital I visited. The temple I go to. The neighbourhood I drove through that one time when I was looking at flats. Every trip, every errand, every deviation from my routine. Google Maps does not just give you directions. It remembers every direction it ever gave you.

YouTube History. Every video I have ever watched. Every video I searched for. Every video I watched for two seconds and skipped. The algorithm does not forget the skip. It remembers what held my attention and what did not, and it uses both to build a model of my interests, my beliefs, my emotional state.

Voice and Audio. Here is one that genuinely unsettled me. Google had audio recordings of my voice. Every time I said “OK Google” or used voice search, it was recorded. I played back a few. Hearing my own voice from three years ago, asking for directions to a place I had forgotten about, was surreal. I did not know these recordings existed. I had technically consented to them. I had no memory of consenting.

Gmail. Every email I have sent and received for over a decade. Conversations with friends, family, clients, doctors, banks, lawyers. Google stopped scanning Gmail content for ad targeting in 2017. But the data is still there. Sitting in the archive. Processed by systems that power features like Smart Reply and inbox categorisation, extracting meaning from your conversations whether or not it is used for advertising.

Google Pay and Purchases. Transaction records. What I bought, where, when, and how much I paid. Subscriptions. Recurring payments. One time purchases. A financial diary I never intentionally kept.

Google Fit. Step counts. Heart rate data. Sleep patterns. Exercise routines. The physical dimensions of my daily life, tracked by sensors in my phone and watch.

Chrome Browsing History. Every website I have visited through Chrome. Every article I have read. Every page I lingered on. Every tab I left open. The browsing history alone is a psychological profile.

Contacts and Calendar. Who I know. When I meet them. How often. Where. The structure of my social and professional life, mapped in calendar entries and contact lists.

What 147 Gigabytes Means

To be fair, Google does give you tools to manage this. You can pause location history, delete voice recordings, set auto delete timers, and review your data at myactivity.google.com. The fact that Google Takeout exists at all  letting you download everything they have  is more transparency than many companies offer. But that is precisely what makes the exercise so revealing.

Here is what struck me hardest. This is not data I actively gave Google. I did not sit down and write a diary for them. Most of this data was generated passively, as a byproduct of living a normal life in 2026. I searched for things because I needed answers. I went places because I had somewhere to be. I watched videos because I was curious or bored or stressed. I sent emails because that is how the world communicates.

At no point did I sit down and think, “I would like to create a comprehensive record of my movements, interests, relationships, health concerns, financial behaviour, entertainment preferences, and inner questions, and hand it to a company in California.” But that is exactly what I did. Over twelve years, one search at a time, one location ping at a time, one email at a time.

And here is the thing. Google is not the only company with data on me. Apple has its share. Microsoft has another. Meta, Amazon, Swiggy, Zomato, PhonePe, my bank, my telecom provider, the half dozen apps I downloaded, used once, and forgot to delete they all have their piece. My 147 gigabytes from Google is just the portion that one company was willing to let me download. The total picture, spread across every platform I have ever touched, is larger than I can comprehend.

Now multiply this by 886 million. According to the IAMAI Kantar Internet in India 2024 report, that is how many active internet users India has. Every single one generating data at some scale. Not all at 147 gigabytes, of course  usage varies enormously between a heavy smartphone user in Bangalore and someone accessing the internet on a shared device in rural Bihar. But collectively, the digital portrait is not just big. It is a civilisational record. And almost none of us have any idea what is being done with it.

Where Your Data Goes When It Leaves You

Here is the part that made me put down my chai and stare at the wall for a while.

Your data does not just sit in Google’s servers, waiting for you to search for a restaurant. It moves. It feeds systems. It trains models.

In February 2024, Google signed a deal with Reddit worth 60 million dollars per year for access to Reddit’s content for AI training. OpenAI signed a similar deal in May 2024, estimated at around 70 million dollars annually (the exact figure has not been publicly confirmed by either company). Reddit disclosed in its IPO filing that it had entered into data licensing contracts worth 203 million dollars. The product being sold? Posts and comments written by ordinary people who thought they were participating in a community discussion, not generating training data for Silicon Valley’s next product.

When Stack Overflow made a similar deal to license its developer Q&A content for AI training, some users fought back. They tried to edit or vandalize their own posts in protest, replacing highly rated answers with objections to the deal. Some invoked the GDPR’s right to erasure as legal justification. Stack Overflow suspended those users. The message was clear: you wrote it on our platform, it is ours now.

This is the economy your data lives in. You create it. Platforms capture it. AI companies buy it. And the AI models that result from this process — the chatbots you talk to, the recommendation engines that shape what you see, the credit scoring algorithms that decide your financial future — are all built on the collective digital exhaust of billions of people who were never meaningfully asked.

As of October 2025, according to a legal tracker maintained by chatgptiseatingtheworld.com, there were at least 51 active copyright lawsuits against AI companies. The New York Times sued OpenAI. Getty Images sued Stability AI. Authors, artists, musicians, all asking the same question: who gave you the right to use this?

It is worth noting that the legal picture is not one sided. In November 2025, an English High Court ruling in Getty Images v Stability AI found that training AI models on copyrighted images was not, in itself, copyright infringement under UK law. The question of whether AI training constitutes “fair use” under US law remains unresolved, with no court decision expected before mid 2026 at the earliest. The legal battles are far from settled, and the outcomes will shape how this data economy evolves.

But those are companies and creators with lawyers. What about you? What about the hundreds of millions of Indians whose search queries, location trails, transaction records, and voice recordings are feeding into AI systems that will make decisions about their lives for years to come?

The Experiment I Am Asking You to Try

I want you to do what I did.

Go to takeout.google.com. Request your data. Wait for the download. Then open it. Not to read every email or trace every location — that would take months. Just look at the categories. Look at the file sizes. Look at how far back it goes.

Then sit with the question that I am still sitting with.

You generated this data by living your life. By asking questions, going places, talking to people, buying things, watching things, reading things. You did not set out to build a surveillance archive. But that is what exists. One hundred and forty seven gigabytes of my life. Probably more of yours.

This data is training AI models you have never heard of, built by companies you have never interacted with, for purposes that did not exist when the data was generated. Your searches from 2015 are informing a language model released in 2026. Your location data from last Tuesday is shaping a recommendation algorithm that will decide what a stranger in another city sees tomorrow.

This data is training AI models you have never heard of.

And nobody asked you. Not really.

I’m have written a book about exactly this  how AI and automated systems make decisions about your life, where accountability disappears, and what we can do about it. If you want to know morea about this book or order a copy, you can do it here: https://akshaywalimbe.com/beyond-bias/