# How Private Is Your Browser? We Ran the Same Tests Across 5 Browsers
I ran our browser fingerprint test and tracking scanner on five different browsers. Same machine. Same network. Same test page. The results were not the same.
Here is the raw data.
Methodology
I tested Chrome 125, Firefox 126, Safari 17.5, Brave 1.68, and Tor Browser 13.5. Each was run fresh with default settings. No extensions added. No settings changed except where the browser ships protections turned on by default. The test page was a standard news site with embedded ads, analytics scripts, and social media widgets.
All tests ran through our Browser Fingerprint tool and Tracking Scanner at duellinghares.com/tools/.
Google Chrome 125
Entropy score: 17.8 bits
Trackers detected: 23
Canvas fingerprinting detected: Yes
WebGL fingerprinting detected: Yes
Chrome leaked 23 trackers on a single page load. The entropy of 17.8 bits means Chrome could identify one user out of roughly 200,000.
Notable findings: Chrome exposed 73 system fonts during the canvas test. The WebGL renderer was reported as your actual GPU model. No built-in tracker blocking exists unless you enable “Do Not Track,” which trackers ignore anyway. Every extension, every custom font, every monitor resolution difference adds to the entropy.
Mozilla Firefox 126
Entropy score: 15.2 bits
Trackers detected: 8
Canvas fingerprinting detected: Yes
WebGL fingerprinting detected: Yes
Firefox ships Enhanced Tracking Protection in Standard mode. It blocked 15 of the 23 trackers Chrome found. The remaining 8 were third-party scripts ETP allows by default. Strict mode drops the count further but can break some sites.
Notable findings: Firefox still exposes a real canvas hash and a real WebGL hash. That means fingerprinters can still identify you. The entropy is lower than Chrome because Firefox has fewer total users. But the browser itself does not fake or randomize any fingerprint data. You need about 15 bits to match a user among 32,000. Firefox gets you there.
Safari 17.5
Entropy score: 13.1 bits
Trackers detected: 6
Canvas fingerprinting detected: Partial (randomized)
WebGL fingerprinting detected: Yes
Safari ships Intelligent Tracking Prevention which blocks cross-site tracking by default. It also randomizes the canvas output on each page load. Fingerprinters can still detect canvas API usage. But the hash changes every time, so the data is useless for matching across sessions.
Notable findings: The 6 trackers we found were all first-party analytics scripts. Safari blocks third-party trackers aggressively. The WebGL fingerprint still leaks through. Safari reports the real GPU model. Font enumeration is partially restricted. Safari hides many system fonts from JavaScript, but not all of them.
The entropy of 13.1 bits is lower than Chrome and Firefox. Safari users look more alike. That is Apple’s design. They reduce the number of data points available to scripts.
Brave 1.68
Entropy score: 8.4 bits
Trackers detected: 4
Canvas fingerprinting detected: Yes (farbled)
WebGL fingerprinting detected: Yes (farbled)
Brave ships Shields Up by default. It blocks scripts, trackers, and fingerprinting attempts before they load. The 4 trackers we detected were first-party. Brave calls this “farble” protection. Canvas and WebGL APIs still respond to JavaScript but return slightly randomized data. The randomization is consistent within a single session so pages still render correctly. Between sessions the values differ.
Notable findings: Brave also spoofs the WebGL renderer string. It returns a generic value instead of your actual GPU. The user agent is modified by default. Screen resolution data is rounded to reduce uniqueness. The entropy score of 8.4 bits means Brave can only identify one user out of about 330. That is a massive improvement over Chrome’s 200,000.
Brave blocks fingerprinting by reducing the signal, not by eliminating features. Your browser still works. Trackers just get bad data.
Tor Browser 13.5
Entropy score: 5.1 bits
Trackers detected: 2
Canvas fingerprinting detected: No
WebGL fingerprinting detected: No
Tor Browser is a hardened browser built specifically to make every user look identical.
Notable findings: The canvas API returns a blank white image. WebGL is completely disabled. Screen resolution is locked to a standard 1000×900 window and rounded with letterboxing. The browser window reports a fixed content window size regardless of your actual window dimensions. No browser extensions leak. The user agent is the same for all Tor Browser users on the same version.
The entropy of 5.1 bits means Tor can only distinguish one user out of about 34. That is the entire point. When 10,000 Tor Browser users visit a site, the fingerprinting script sees roughly 300 distinct fingerprints. Not 10,000. The 2 trackers we detected were first-party scripts that loaded before Tor’s protections kicked in.
Chrome and Firefox expose your real fingerprint to any site that asks. Safari and Brave reduce the signal. Tor eliminates it almost entirely.
Tor Browser wins on pure privacy. But Tor has tradeoffs. It is slow. Many sites block Tor exit nodes. Some features like WebGL are disabled entirely.
For daily use, Brave is the practical winner. It blocks 19 more trackers than Chrome. It fakes fingerprint data without breaking websites. It loads faster because it blocks so many scripts. The privacy gain costs you nothing in usability.
If you do not want to switch browsers, use Firefox with fingerprinting resistance enabled. Go to about:preferences#privacy, set Enhanced Tracking Protection to Strict, and enable “Resist Fingerprinting” in about:config. That gets you close to Brave’s protection without leaving Firefox.
Safari is fine if you are already in the Apple market. It is better than Chrome or default Firefox. But it still leaks WebGL data through, which Brave does not.
Chrome is the worst option for privacy. It is the most popular. It is also the most trackable. If you are using Chrome without any extensions, every ad network on the internet can fingerprint you with high confidence.
Test Your Own Browser
Every browser is different. Every setup adds variables. Extensions, fonts, screen resolution, operating system. These all change the numbers.
Go to duellinghares.com/tools/ and run the tests yourself. Our Browser Fingerprint tool shows your exact entropy. The Tracking Scanner shows what trackers are running on the pages you visit. Both run completely in your browser. Nothing is sent to a server.
The numbers do not lie. Run the test. See where you stand.
*Victor Kane, Security & OSINT Analyst*