It was discovered that AliExpress implemented discreet background audio processes within the browser with the aim of identifying and monitoring visitors' computers.
This practice came to light after a developer noticed an anomaly during the automatic switching of a Bluetooth headset connection while browsing the e-commerce platform.
According to TechSpot, the platform maintains an active audio stream in the browser, even when the volume is set to zero. This is enough to keep the computer's audio system occupied, causing interference with connections to other devices.
The methodology employed is called fingerprinting and can serve platforms as an alternative to conventional cookies for fraud prevention and suspicious activity detection. However, this technique is criticized for operating silently and collecting usage data without obtaining user permission, similar to what happens with microphones.
Following the revelation of this incident, developers of browsers such as Brave and Firefox stated on social media that their software includes tracking script blockers, like those used by AliExpress.
The developer in question was using a Bluetooth headset with multi-point functionality, which allows the accessory to remain connected to two devices and switch between them. However, upon accessing AliExpress, he observed that the headset remained connected to the computer even when he intended to transfer the audio to his cell phone; this behavior ceased as soon as the store tab was closed.
By examining the source code of the e-commerce homepage to understand the headset's operation, the developer identified that the site was using the Web Audio API—a standard browser feature—to generate audio processing graphs.
This process occurred without sound, but since the script connects to the operating system's sound output, the machine's audio route remains occupied in the background. This integration prevents control tools, such as the browser tab's native mute function, from having any effect.
AliExpress's mechanism builds a unique identifier for the machine by combining various signals. The site transmits a standardized audio signal and measures variations in how the equipment processes and returns the sound, which differs depending on components such as the processor, sound card, sound drivers, and browser version.
TechSpot reported that the data collection also includes metrics that allow for the creation of an extremely detailed profile of each computer, covering various hardware and software characteristics.
