Abstract
Whistleblower laws protect individuals who inform the public or an authority about governmental or corporate misconduct. Despite these laws, whistleblowers frequently risk reprisals and sites such as WikiLeaks emerged to provide a level of anonymity to these individuals. However, as countries increase their level of network surveillance and Internet protocol data retention, the mere act of using anonymizing software such as Tor, or accessing a whistleblowing website through an SSL channel might be incriminating enough to lead to investigations and repercussions. As an alternative submission system we propose an online advertising network called AdLeaks. AdLeaks leverages the ubiquity of unsolicited online advertising to provide complete sender unobservability when submitting disclosures. AdLeaks ads compute a random function in a browser and submit the outcome to the AdLeaks infrastructure. Such a whistleblower's browser replaces the output with encrypted information so that the transmission is indistinguishable from that of a regular browser. Its back-end design assures that AdLeaks must process only a fraction of the resulting traffic in order to receive disclosures with high probability. We describe the design of AdLeaks and evaluate its performance through analysis and experimentation.
Introduction
In the current landscape of pervasive network surveillance, traditional anonymity tools like Tor or encrypted SSL/TLS channels often fall short for high-risk whistleblowing. While these tools protect the content of a message, they fail to hide the metadata—the fact that a specific user is communicating with a known whistleblowing platform. This "signal of intent" can be sufficient for state actors to initiate investigations even if they cannot read the message itself.
To address this, the AdLeaks system aims for unobservability: a state where an adversary cannot distinguish between a whistleblower and a regular internet user. The core innovation lies in leveraging the massive, unsolicited traffic generated by online advertising to provide a cover for sensitive data transmissions.
The AdLeaks Architecture
AdLeaks operates by integrating with existing web advertising ecosystems. The system consists of four primary components: the Ad Network, the Whistleblower's instrumented browser, the Backend Aggregators, and the Decryptor.
The process begins when a regular user visits a partner website, such as a news outlet. An AdLeaks ad is served, which executes JavaScript to encrypt a "0" (representing noise) using a public key. This creates a massive baseline of "white" ciphertexts. A whistleblower, using a specialized browser, identifies these ads and replaces the "0" with a chunk of actual disclosure data ("gray" ciphertext). Because the encryption is performed locally and the resulting ciphertexts are indistinguishable, an observer cannot tell which user is sending noise and which is sending data.
Backend Infrastructure and Homomorphic Aggregation
To manage the enormous volume of traffic generated by millions of users, AdLeaks employs the Damgård-Jurik (DJ) homomorphic encryption scheme. This allows intermediate "Aggregator" servers to combine multiple ciphertexts without decrypting them. Mathematically, the product of two ciphertexts corresponds to the sum of their plaintexts.
This property enables a "Tree-Decryption" optimization. If the aggregate sum at the root of a processing tree is zero, the system can discard the entire branch of thousands of messages immediately. Only when a non-zero sum is detected (indicating a disclosure chunk or a "black" ciphertext collision) does the system descend further into the tree. To mitigate collisions, the system utilizes Fountain Codes, allowing for the reconstruction of the original file even if some data chunks are lost or corrupted during aggregation.
Performance and Scalability
The system was evaluated using empirical data from WikiLeaks archives to model disclosure sizes and Alexa traffic statistics to estimate reach. Benchmarks conducted on the EmuLab network testbed demonstrate significant scalability.
A single aggregator can process approximately 8,000 requests per second. With a standard 18 Mb/s connection, the decryption backend can support over 51,000 concurrent whistleblowers. While the security guarantees are high, there is a trade-off in latency: a 2MB file might take roughly 21 days to submit if the whistleblower limits their transmissions to 50 per day to maintain a low profile and avoid traffic spikes.
Conclusion
AdLeaks presents a paradigm shift in secure communications by hiding disclosures within the "noise" of the internet. By utilizing the unsolicited nature of ads, it removes the need for users to "opt-in" to a suspicious cover-traffic network. While challenges remain regarding submission latency and potential domain blocking by adversaries, AdLeaks provides a robust framework for achieving sender unobservability in an era of global surveillance.
Summary updated