The spy in your office: why Chrome is an unacceptable risk and how to practise sovereign digital hygiene

If you glance at the screen of any office worker today, you will see a common denominator: a circular red, yellow and green icon. Google Chrome holds more than 65% of the global market share. It is fast, convenient and, apparently, free.

But in enterprise architecture, there is no such thing as free; you pay with metadata.

Using Chrome as the main window onto your company's CRM, invoicing, bank accounts and your clients' email is like installing a surveillance camera in the boardroom and handing the remote control to an advertising giant. Today at Oksigenia we are not going to talk about “incognito mode” (a cosmetic patch), but about browsing architecture, profile segregation and true data sovereignty.

“Your browser is the border between your company's internal network and the outside world. Letting an advertising company manage that border is architectural negligence.”

The single-ecosystem trap

The problem with Chrome is not that it is bad software; technically, the Chromium engine is a brilliant piece of engineering. The problem is its business model: constant telemetry. Chrome is designed to tie your identity, your history, your extensions and your passwords to a centralised profile.

When your whole office runs on this model, you are giving away business intelligence. What hours do you work? Which SaaS platforms do you use? Which banks do you visit? All that telemetry travels to other people's servers.

To regain control, changing browser is not enough; you have to change strategy. This is where digital segregation comes in.

The Firefox dilemma and real isolation

The historical alternative to Chrome has always been Firefox. However, there is a heated (and justified) debate in the technical community. Mozilla has recently made some highly controversial corporate decisions: telemetry built in by default, questionable search deals and advertising experiments in the browser.

So why are the most secure browsers in the world, such as Tor Browser or Mullvad Browser, built on Firefox?

Because its engine (Gecko) allows a level of isolation that Chromium (the basis of Chrome, Edge and Brave) can barely dream of. Firefox has a native architecture of containers (Multi-Account Containers).

// The logic of containers versus cross-site cookies:
if (navegador == "Firefox_Containers") {
    // The bank tab does not know the social media tab exists
    aislar_sesion(pestaña_A) != aislar_sesion(pestaña_B);
    rastreo_cruzado = "BLOQUEADO";
}
// In a traditional browser, all tabs share the same "memory".

A container lets an employee keep the admin panel session (WordPress, n8n, Proxmox) open in one tab and their email in another, with no cookie or tracker communication whatsoever between them.

Brave and the Chromium challenge

For the day-to-day work of the “average employee”, imposing a heavily configured (hardened) Firefox can create friction, as some poorly designed websites may break. The painless transition is called Brave.

Brave uses the same engine as Chrome (ensuring that 100% of websites work), but rips Google's telemetry out at the root and blocks trackers at network level (Shields).

However, Brave inherits an architectural problem from Chromium: it has no real containers. Although they have been on its technical roadmap for some time, implementing strict tab isolation in an engine Google designed to share data is a titanic challenge. For now, it relies on creating separate “profiles”, which is more cumbersome for the user.

The Oksigenia standard: tactical compartmentalisation

In our services division, when we deploy Linux workstations or audit offices, we are not looking for the “perfect browser” but for the right architecture. We apply the principle of compartmentalisation:

1. The “workhorse” (Brave): For employees' general browsing, searches and office work. It is compatible with everything, but with its shields up. It blocks ads and trackers and sends no telemetry to Mountain View.
2. The “admin bunker” (Mullvad Browser / Firefox Hardened): We use strictly configured (and segregated) Gecko-based browsers only to access critical infrastructure: VPS servers, databases or payment gateways. This browser is not used to search Google or to visit social media.
3. Mandatory containers: If the workflow requires keeping several sessions open, we configure environments where third-party tools are confined to their own digital “jail”.

Giving up Chrome is not the end goal; the goal is for your company to decide which data leaves the office and which stays. And that decision starts in the address bar.