What Glassbox Actually Shows You
Glassbox is designed to reveal a browser’s unique fingerprint using over 30 probes. This includes canvas rendering, WebGL data, font libraries, WASM features, API matrices, and cross site login states. The tool pings a public geolocation API, but every other piece of processing happens locally on the user’s machine.
Unlike existing tools such as AmIUnique or the EFF‘s Cover Your Tracks, Glassbox goes further by providing a detailed breakdown alongside an estimated identifiability score, giving users a clearer sense of exactly how exposed their setup really is.
Built With Help From Claude Code
Dale, a solo entrepreneur and security engineer, used Claude Code to refine his ideas while building the tool. He said in a Hacker News thread that AI tools made it “much easier to polish ideas,” and that he wanted a single place to see every fingerprinting method tested against his own browser.
The idea traces back to Dale learning about silent sawtooth waves, a technique used by fingerprinting code on Alibaba’s AliExpress to identify browsers through audio signals, as previously reported.
Some Browsers Blend In Better Than Others
Testing revealed sharp differences depending on which browser you use.
- Chrome, Edge, and Safari sessions were effectively unique, meaning almost no other user shared the same fingerprint
- Firefox came in at 89 percent identifiability, shared with roughly 1 in 681 million users
- Tor Browser scored 56 percent, shared with about 1 in 408 thousand users
Dale’s own Chrome session showed an identifiability of roughly 1 in 7.6 billion browsers, essentially unique among the entire global population using the internet.
Why the Number Isn’t the Whole Story
Dale was careful to note that the identifiability score is an estimate, not a hard measurement. “It sums published per signal entropy, discounts your browser masks, and caps at the ~33 bits needed to single out one person on Earth,” he explained. Since Glassbox never sends data to a server, it has no way to measure true rarity against a real world dataset.
His advice for reducing exposure comes down to one core idea: blend in rather than stand out. “The single highest impact move is to use a browser that puts you in a large, identical crowd,” Dale said. Ironically, a heavily customized browser setup, extensions, unusual settings, tweaked configurations, often makes a user more identifiable, not less.
He recommends using a VPN or Tor and closing the WebRTC leak, a common gap that can expose a real IP address even behind a VPN. Glassbox includes a full page of suggestions for anyone looking to reduce their own fingerprint.
Follow Hashlytics on Bluesky, Facebook, LinkedIn , Telegram and X to Get Instant Updates