Om sesjonen
Mobile applications have always shared extensive data with third-party advertising and analytics companies. When I revisited privacy research conducted several years ago, I found the landscape had shifted: where plaintext payloads once appeared in a proxy, we now see opaque encrypted blobs even after TLS is stripped. Application-layer encryption has become the new barrier between researchers and the data.
This talk demonstrates how to break through that barrier without ever knowing the encryption key. Using Perfect365, a beauty app with over 100 million installs, as a case study, we walk through the complete methodology: identifying double-encrypted traffic in a proxy, tracing encryption routines through heavily obfuscated Android bytecode, and instrumenting those routines with Frida to intercept data at the point it enters the cipher, before the key is applied.
The technique turns any application-layer cipher into a transparent window. We reveal what ad SDKs actually transmit across the real-time bidding supply chain: auction payloads, cross-publisher identity tokens, device fingerprints, and granular ad lifecycle telemetry. This data is encrypted precisely because it was never meant to be seen.
We also discuss how AI-assisted static analysis has fundamentally changed the pace of this research. Navigating hundreds of thousands of lines of obfuscated SDK code that would previously have taken weeks now takes hours, lowering the barrier for privacy researchers significantly.
Attendees will leave with a reusable Frida methodology applicable to any app using AES, RSA, or hybrid encryption, and a clearer picture of the data economy operating invisibly inside the apps on their devices.