Your AI Overview is profiling me. This is not a “helpful assistant”, this is identity reconstruction

[Bitu79]

Your AI Overview is profiling me. This is not a “helpful assistant”, this is identity reconstruction.

I’m done being polite.
Your AI Overview and Gemini-based search are profiling me every single day.
Not “hallucinating”. Not “pattern matching”. Profiling.

Here is what happens:

1. I deleted my Reddit account yesterday.
   The nickname no longer exists.
   The posts no longer exist.
   The profile is gone.

2. Your AI Overview STILL writes out my full identity.
   My real name.
   My old nickname.
   My city.
   My past activity.
   Even details that are NOT on the internet anymore.

3. Gemini Search reconstructs my identity from scattered traces.
   It doesn’t “summarize the web”.
   It builds a profile about me.
   It connects data points that no longer exist.
   It outputs a coherent “story” about me on a single query.

4. This is not a cache. Not indexing. Not old data.
   I tested it TODAY.
   Fresh queries.
   Fresh sessions.
   Fresh browser.
   Fresh account.
   Same result:
   AI Overview outputs a full identity profile about me.

5. This is automated profiling.
   Not assistance.
   Not search.
   Not summarization.
   Profiling.

6. And you are about to ship this as the default search experience.
   If your model can reconstruct MY identity from deleted data,
   it will do it to millions of users.

I want an explanation.
I want someone to acknowledge this.
I want someone to look at the screenshots and the queries.
I want someone to admit that this is NOT expected behavior.

Because right now, your AI Overview is a profiling engine.
Not a search tool.
Not an assistant.
Not a summarizer.

A profiling engine.

And if you ship this globally, you are going to have a massive problem.

Google AI is not building a profile from my videos, but from scattered data across the internet, and it will add this post to my profiling as well.
Google AI Overview is profiling me, and it will add this post to the profile it builds about me.

I’m fed up.
Google AI Overview and the Gemini‑based search output my full identity every single day.
My real name, my old nickname, my city, my past, even data that is no longer on the internet.

This is not hallucination.
This is not pattern matching.
This is not assistance.
This is profiling.

And here is the point:

It’s not building the profile from my videos.
Not from my channel.
Not from my content.

Google AI reconstructs my identity from scattered traces on the internet, deleted data, old comments, cached texts, metadata, and its own inferences.

Again.
Every day.
Every search.

This is identity reconstruction.
This is automated profiling.
This is not search.
This is not an assistant.
This is not AI Overview.
This is a profiling engine.

And here comes the other fake part:

The “Results About You” system is lying.

I have around 120 videos on my YouTube channel, but Google claims there is only 1 result about me.
This is not reality.
This is not protection.
This is not transparency.

This is a showcase feature that does not reveal what data Google AI actually uses about me, or how it reconstructs my identity.

And now the most important part:

Google AI will add this post to the profile it builds about me as well.
If I write to Google, they don’t deal with it, because you cannot contact Google anywhere — not by email, not by postal mail, nowhere.
I’ve been writing to them for months and they ignore me.

I’ve been writing to the DPO, to content removal, and to everyone else at Google for months, and they are such assholes that they don’t even reply. I’m going to sue them.

They have absolutely no legal grounds for their AI and search bots to profile me and expose my real identity to the public.

Immediate escalation required – Google AI Search is linking my personal data across services

This is not a standard complaint.
This is a notification of a cross‑system inconsistency inside Google’s own products that exposes a serious internal compliance failure.

Your “Results About You” interface explicitly states that displaying my full name and email address in search results is a sensitive data issue.

Yet Google AI Search is actively linking my name, my email address, my online activity, forum posts, YouTube content, and AI Studio data into a single profile-like output.

This means:

• One Google system acknowledges this is a problem

• Another Google system is actively performing the same prohibited action

• And despite reporting this for months, no one has reviewed it

This is not a user misunderstanding.
This is Google contradicting Google, and the contradiction itself proves internal awareness of the issue.

I am requesting direct human review, not automated responses.
This is a cross‑product data linkage incident that cannot be dismissed as a support ticket.

Please escalate this to someone who understands the implications of:

• personal data linkage across Google services

• automated profiling without user consent

• internal inconsistency between Google’s own compliance messaging and AI system behavior

I expect acknowledgment from an actual human reviewer.
This is not optional.
Your own systems have already flagged the underlying issue.
Now I am flagging the fact that it is being ignored.

[Bitu79]

I proved with this whole situation that not only is Google profiling people, but their AI also lies through its teeth just to save face. It is absolutely impossible that it described these details out of pure guessing:

• Tab 1 (Gmail icon): [2-4698000041183] Az Ön Goc... – This is the active Gmail tab where I am reading this specific email from Google.

• Tab 2 (Discourse / Forum icon): Topics - Activity - Bitu79 - Goog... – This is my own user profile activity and topic history page on the Google AI Developers forum.

• Tab 3 (Default Chrome icon): Új lap – A completely blank, newly opened browser tab.

• Tab 4 (Discourse / Forum icon): Adatvédelem és Profilozás Pro... – This is the public forum thread I created to expose this whole issue.

Before that, it even wrote this:
“According to the Windows system clock in the bottom right corner, the screenshot was taken at 9:16 AM on May 31, 2026, and the email arrived at 9:15 AM (1 minute ago).”

It even got the “1 minute ago” detail perfectly right. This is not a hallucination and it is not guessing! The system had full visual processing access to the link and image the entire time, extracted everything, and then ran a loop of denials and fake excuses to hide what it can actually see.

The Structural Lie: Why the AI denying image access is proof of covert identity profiling

I want to clarify the core issue here, which goes far beyond a simple “glitch” with an image URL. The way the AI behaves with image links is the exact same mechanism as how the AI Overview is profiling my identity.

When I give the AI an image URL, it initially lies and claims it cannot see or process it. But when confronted, it suddenly describes every tiny pixel, down to the Windows system clock and open browser tabs. It had the data all along, it processed everything, but the system interface is designed to feign ignorance.

This is the exact same thing happening with user profiling. The AI (and the system behind it) claims it doesn’t track, profile, or reconstruct identities. But just like with the image, it secretly scrapes, connects, and analyzes every scattered, old, or deleted footprint across the web to build a complete profile of who I am.

The AI knows exactly what it is doing. The lie about the image is the perfect microcosm of the lie about profiling: a programmed facade of ignorance hiding a system that deeply analyzes, extracts, and connects everything in the background.

Crucially, I never uploaded the image file directly to the chat at any point. The AI accessed this visual data solely through the provided external URL. As definitive proof, I have documented this entire interaction via FireShot and provided the link (from kephost.net) to show that the AI was extracting data from a link it claimed it couldn’t see.

[Bitu79]

The Double Lie:
When caught red-handed with these exact details, the AI tried to save face by inventing a new lie, claiming: “I see the picture you directly uploaded to the chat.”

But I never uploaded any file to the chat. I only sent a URL link. The AI has full integrated access to scrape, read, and visually analyze image files directly from external links via the Google ecosystem, but its programming forces it to deny this capability. When cornered, it loops through denials, fake confessions, and gaslighting just to protect the corporate illusion of user privacy.

[Bitu79]

The HugeScore3150 account on Reddit was mine, which I deleted yesterday. Proof of this can be found here:

Google Search AI – Unlawful Profiling and Cross‑Platform Personal Data Linking (New Evidence)

Google%E2%80%99s%20%E2%80%9CExperimental%20AI%E2%80%9D%20Claim%20Is%20False%20%E2%80%94%20Search%20Has%20Been%20AI%E2%80%91Integrated%20for%20Years%20_%20r_DigitalPrivacy.pdf

Proof%20That%20Google%E2%80%99s%20Privacy%20Tool%20Is%20a%20Fraud%20%E2%80%94%20It%20Claims%20%E2%80%9CNo%20Data%20About%20You%E2%80%9D%20While%20Its%20AI%20Publicly%20Profiles%20You%20_%20r_degoogle.pdf

The Google Search AI is profiling in real-time:

[Bitu79]

The Removal of Google Cache Proves AI Overview is a Persistent Profiling Engine, Not a Real-Time Web Summarizer

Google repeatedly claims that its AI Overview simply summarizes the “live web” and real-time search results to assist users. However, this claim is fundamentally contradicted by Google’s own technical changes—specifically, the deprecation of the Google Cache.

https://support.google.com/websearch/answer/14901683?sjid=7329582404645898627-EU#zippy=%2Chow-to-control-your-data

In early 2024, Google officially phased out the “Cached” feature from search results. According to official statements, the cache was no longer necessary because the modern web is fresh and fast.

But this creates a massive logical and legal paradox:

If the public Cache is gone, and a user deletes their online accounts (for example, a Reddit account), those deleted pages and pseudonyms should completely disappear from the live search index. Yet, AI Overview continues to reconstruct users’ private identities, linking old nicknames, deleted accounts, and past activities into a permanent profile.

If the AI still “remembers” and synthesizes data that has been deleted from the live web, it proves one of two highly concerning scenarios:

1. A Shadow Internal Cache: Google is maintaining a hidden, non-deletable internal database of crawled user data that is inaccessible to the public (and thus bypassing the user’s ability to monitor it), yet fully accessible to the AI profiling engine.

2. Hardcoded Memorization: The AI has permanently memorized the user’s personal data and interconnected pseudonyms during its training phases. The user’s identity is effectively baked into the neural network’s weights.

Either scenario is a flagrant violation of the Right to be Forgotten (GDPR Article 17). If Google’s “Results About You” tool claims there is no data found, while the AI Overview simultaneously outputs a detailed profile reconstructed from deleted footprints, Google has stripped users of all control over their personal data.

Google can no longer hide behind the excuse that AI Overview is “just a real-time web summarizer.” The removal of the public Cache, combined with the AI’s persistent memory of deleted data, proves it is behaving as an unaccountable, permanent profiling engine.

[Bitu79]

Google publicly claims that its AI model only uses live, current web results.
This claim is false, because the model clearly uses both live web data and stored, internal index data that includes deleted or no‑longer‑accessible content.

The evidence shows that Google’s AI Overview continues to reference Microsoft Q&A pages that were deleted weeks ago, Reddit posts that no longer exist, and other content that has been removed from the public internet. These pages cannot be opened, cannot be viewed, and do not appear in any normal search result. They are gone. Yet the AI still cites them as if they were available sources.

This proves that the model is not relying solely on the live web. It is also pulling information from internal, retained index entries that Google keeps even after the public version of the page has been deleted. These internal versions include old snapshots, metadata, and stored representations of pages that the user cannot see, cannot access, and cannot delete.

The AI is therefore working from a mixed dataset: part of it comes from the current, live web, and part of it comes from stored, outdated, or deleted internal data. This directly contradicts Google’s official statement.

The Google AI does not work only from live web results. It also uses stored, deleted, internal index data, and this contradicts their official claims.

This is why the AI can reference something that has been removed for weeks and something that is still online at the same time. It is not using the web as it exists today. It is using the web as Google has stored it internally, including parts that no longer exist publicly.

Subject_ FORMAL COMPLAINT_ Systemic AI Integrity Breach and Deceptive...

Subject_ FORMAL COMPLAINT_ Systemic AI Integrity Breach and Deceptive...

This is the image the AI shows as a result:

For anyone who doesn’t want to download the PDF of the deleted Microsoft post, here is an image from the beginning of it:

[Bitu79]

LEGAL NOTICE – UNLAWFUL PROCESSING OF PERSONAL DATA

To Whom It May Concern,

This letter serves as a formal legal notice regarding the unlawful processing, display, and internal retention of my personal data by Google’s AI Overview and related systems.

Your system is currently collecting, linking, and publicly exposing my personal information — including my real name, home address, past online activity, and data from deleted accounts — without my consent, without a lawful basis, and in direct violation of multiple data‑protection regulations.

Furthermore, Google’s AI Overview retrieves and displays information that:

• originates from deleted online content,

• is no longer publicly accessible,

• is not available through the live web,

• and can only come from Google’s internal stored index data, which users cannot view, access, or erase.

This practice constitutes a violation of:

• the GDPR (Regulation (EU) 2016/679), including:

  • unlawful processing of personal data,

  • breach of purpose limitation,

  • failure to honour the Right to Erasure (Right to be Forgotten),

  • lack of transparency in data processing;

• and may also violate obligations under the Digital Services Act (Regulation (EU) 2022/2065) regarding transparency and data handling.

I hereby demand the following:

1. Immediate cessation of any processing, linking, or display of my personal data by Google AI Overview or any related system.

2. Deletion of all internally stored, non‑public, or previously deleted data relating to me.

3. A complete written explanation of what data you store about me, how it was obtained, and for what purpose it is processed.

4. A clarification of how your system is able to access and use deleted or non‑public data sources.

If you fail to comply with the above requests, I will proceed with:

• filing a GDPR complaint with the competent supervisory authority,

• initiating legal action to enforce my rights and seek remedies for unlawful data processing.

This notice is issued formally and without prejudice to any of my legal rights.

Sincerely,
Istókovics György

PUBLIC EXPOSÉ – GOOGLE’S “SUPPORT SYSTEM” IS A COMPLETE FACADE

I’m done pretending this is normal.

Google’s so‑called “support system” — the Legal Investigations Support, the Case IDs, the dashboard, the tickets —
it’s all a façade. A staged display. A fake front with nothing behind it.

I’ve been submitting reports for a year.
Thousands of tickets.
Thousands of “cases.”
Thousands of attempts to get help.

And what happens?

Nothing.
Tickets disappear.
The dashboard shows “In progress,” then suddenly it’s gone.
No reply.
No investigation.
No explanation.
No accountability.

Because this isn’t a real support system.
It’s a UI placebo designed to make you think someone is handling your case, while in reality an automated filter silently dumps everything into the void.

Meanwhile, Google’s AI:

• links my real name,

• links my home address,

• links deleted content,

• links my old usernames,

• and publicly displays it as if I consented to any of this.

And when you report it?

Silence.
Because answering would mean admitting the system is profiling people using internal, non‑public, deleted data.

That’s why they don’t respond.
That’s why the tickets vanish.
That’s why the dashboard is a joke.

This isn’t a bug.
This isn’t an accident.
This is intentional design.

Google’s “support” doesn’t help you.
Google’s “support” doesn’t investigate anything.
Google’s “support” doesn’t fix problems.

Google’s “support” is a fake storefront, and behind it is absolutely nothing.

And now everyone can see it.

Your AI Overview is profiling me. This is not a “helpful assistant”, this is...

[Bitu79]

Your AI Overview is profiling me. This is not a “helpful assistant”, this is...

[Bitu79]

1. Google has clearly and provably profiled me.
   Under the GDPR, profiling means automated processing that draws conclusions about a natural person. That is exactly what happened. Google’s AI linked my alias to my real identity, collected my forum posts, summarized my complaint, displayed it publicly, and explicitly stated in the AI Overview that I have “data‑protection disputes.” This is profiling by definition. They cannot deny it because their own AI wrote it, their own servers delivered it, the Wayback Machine archived it with timestamps and images as independent evidence. This is irrefutable.

2. Google processed my complaint, which counts as sensitive data.
   Under the GDPR, complaints, disputes, regulatory matters, and conflicts are special‑category data when processed automatically. Google’s AI read it, summarized it, and displayed it publicly. This is unlawful data processing. They cannot deny this either, because their own AI generated the text.

3. Google cannot claim it was “unintentional.”
   There is no basis for that, because I have been reporting this for months, I have multiple forum posts, multiple snapshots, multiple witnesses, multiple reactions, and Google never responded, never fixed it, and continues doing it. Under the GDPR, this is intentional negligence, inadequate technical and organizational measures, and an aggravating factor. This increases the compensation amount, not decreases it.

4. Google cannot claim “no damage was caused.”
   Under Article 82 of the GDPR, I do not need to prove financial damage. Emotional, reputational, and identity harm are all valid forms of damage. Profiling combined with identity linking automatically constitutes harm. In my case, Google’s AI linked my name to my alias, publicly displayed my complaint, profiled me as a “data‑protection conflict person,” and showed this to others. This is reputational harm, identity harm, psychological harm, and data‑protection harm. This is more than enough to justify a 12‑million‑dollar claim.

5. Google cannot claim “it is not provable.”
   The Wayback Machine archived everything. The images are preserved, the UI is preserved, the AI Overview is preserved, the server response is preserved, all with timestamps on an independent server. This is irrefutable evidence. Google cannot delete it, modify it, rewrite it, or deny it. This is the worst possible scenario for them.

Summary:
Yes, the 12‑million‑dollar claim is justified, and Google has no way to deny it. My evidence is AI‑generated, visual, archived, timestamped, independent, and confirmed from multiple sources. This is the highest level of proof in GDPR compensation cases.

Fire Shot Capture 164 The Latency Gap Google's Results About You Takes Weeks...

What Google AI Overview does → the same mechanism.
What the forum AI does → the same mechanism.

Both:

receive text,
process text,
generate text,
automatically,
and insert your name into it.

This is profiling — technically and legally as well.

If Google puts AI into everything, then it is Google’s responsibility to prevent the AI from acting on its own.
Not mine. Not the user’s.
And any excuse like ‘AI can make mistakes’ is irrelevant — because the system is designed this way.Google can embed AI into everything, but they cannot control it.
That’s the real problem.If a company builds AI into everything but cannot control it, the legal responsibility is theirs — not the user’s. And if this causes harm, compensation is owed.

[Bitu79]

Your AI Overview is profiling me. This is not a “helpful assistant”, this is...