AliExpress was found to be fingerprinting visitors by sending inaudible sounds to browsers and reading the resulting audio data, a technique that runs on the site’s homepage via a WebAudio‑based script. Security researcher Matthew Callaghan identified the method, noting that the script creates an oscillator, measures the output, and transmits the frequency data back to AliExpress for tracking purposes.
Key Takeaways
- AliExpress employs a WebAudio‑based fingerprinting method that emits inaudible “sawtooth” waves.
- The script generates a known waveform with an oscillator, captures the browser’s audio response, and sends frequency data to AliExpress.
- Researcher Matthew Callaghan discovered the technique after his phone’s audio stopped playing when the AliExpress homepage loaded.
- The method relies on an outdated fingerprinting approach that sets audio gain to zero, making the sound invisible to users.
| Metric | Value | Why it matters |
|---|---|---|
| Company | AliExpress | Uses the technique on its homepage to uniquely identify browsers. |
| Researcher | Matthew Callaghan | Discovered the hidden audio fingerprinting, exposing a privacy risk. |
| Technique | WebAudio oscillator & analyser | Creates a known waveform, measures the browser’s audio implementation, and reads frequency data. |
What the audio fingerprint does
The script runs as soon as the AliExpress homepage loads. It creates an oscillator that emits a “sawtooth” wave at a frequency beyond human hearing. By setting the audio gain to zero, the sound remains inaudible while still traveling through the browser’s audio stack.
How the WebAudio pipeline is leveraged
First, the oscillator generates a known waveform. The browser’s audio implementation processes this signal, after which an analyser node captures the resulting frequency spectrum. The script then reads the frequency data and sends it back to AliExpress, effectively creating a fingerprint that can differentiate one browser instance from another.
“The oscillator generates a known waveform.”
Discovery and immediate impact
Matthew Callaghan observed that his phone’s audio playback halted the moment the AliExpress page rendered, indicating that the script was actively hijacking the audio routing path. This behavior confirmed that the fingerprinting code was not merely passive but was interfering with the device’s audio output to ensure the measurement could be taken.
Why the method is considered outdated
The approach relies on an older class of fingerprinting that measures how browsers handle inaudible WebAudio signals. Modern browsers have introduced mitigations against more sophisticated tracking vectors, but the technique still works because it exploits fundamental audio processing pathways that remain exposed.
Why it matters
Browser fingerprinting threatens user privacy by enabling persistent tracking without cookies or explicit consent. By embedding the audio measurement directly into the homepage, AliExpress can collect a unique identifier before any user interaction occurs. The discovery highlights the need for browsers to tighten audio API permissions and for sites to be transparent about hidden tracking mechanisms.
Frequently asked questions
What exactly does AliExpress do to fingerprint browsers?
It injects a WebAudio script that generates an inaudible waveform, captures the processed frequency data via an analyser node, and transmits that data back to its servers.
Who uncovered this tracking method?
Security researcher Matthew Callaghan identified the technique while testing the AliExpress homepage.
Does the audio fingerprint affect user experience?
The script sets the audio gain to zero, so users do not hear the sound. However, Callaghan noted that his phone’s audio stopped playing when the page loaded, indicating the script can temporarily disrupt other audio streams.
Is this a new form of fingerprinting?
No, it builds on an older method that uses WebAudio “sawtooth” waves to probe browser implementations, a technique that has been known but is rarely seen in the wild.
Has AliExpress responded to the findings?
No public reaction or corporate response has been reported.
Bottom Line
AliExpress’s use of inaudible‑sound fingerprinting demonstrates that even legacy tracking techniques can resurface on high‑traffic sites, underscoring the ongoing privacy challenges posed by hidden WebAudio scripts and the importance of vigilant browser defenses.
Reporting based on: original source.
Comments
Post a Comment