Real-Time Bidding (RTB) and Cookie Matching (CM) are transforming the advertising landscape to an extremely dynamic market and make targeted advertising considerably permissive. The emergence of these technologies allows companies to exchange user data as a product and therefore raises important concerns from privacy perspectives. In this paper, we perform a privacy analysis of CM and RTB and quantify the leakage of users' browsing histories due to these mechanisms. We study this problem on a corpus of users' Web histories, and show that using these technologies, certain companies can significantly improve their tracking and profiling capabilities. We detect 41 companies serving ads via RTB and over 125 using Cookie Matching. We show that 91% of users in our dataset were affected by CM and in certain cases, 27% of users' histories could be leaked to 3rd-party companies through RTB. We expose a design characteristic of RTB systems to observe the prices which advertisers pay for serving ads to Web users. We leverage this feature and provide important insights into these prices by analyzing different user profiles and visiting contexts. Our study shows the variation of prices according to context information including visiting site, time and user's physical location. We experimentally confirm that users with known history are evaluated higher than new comers, that some user profiles are more valuable than others, and that users' intents, such as looking for a commercial product, are sold at higher prices than users' browsing histories. In addition, we show that there is a huge gap between users' perception of the value of their personal information and its actual value on the market. A recent study by Carrascal et al. showed that, on average, users evaluate the price of the disclosure of their presence on a Web site to EUR 7. We show that user's browsing history elements are routinely being sold off for less than $0.0005. Permission to freely reproduce all or part of this paper for noncommercial purposes is granted provided that copies bear this notice and the full citation on the first page. Reproduction for commercial purposes is strictly prohibited without the prior written consent of the Internet Society, the first-named author (for reproduction of an entire paper only), and the author's employer if the paper was prepared within the scope of employment.

We present the results of the first large-scale study of the uniqueness of Web browsing histories, gathered from a total of 368, 284 Internet users who visited a history detection demonstration website. Our results show that for a majority of users (69%), the browsing history is unique and that users for whom we could detect at least 4 visited websites were uniquely identified by their histories in 97% of cases. We observe a significant rate of stability in browser history fingerprints: for repeat visitors, 38% of fingerprints are identical over time, and differing ones were correlated with original history contents, indicating static browsing preferences (for history subvectors of size 50). We report a striking result that it is enough to test for a small number of pages in order to both enumerate users' interests and perform an efficient and unique behavioral fingerprint; we show that testing 50 web pages is enough to fingerprint 42% of users in our database, increasing to 70% with 500 web pages.

We propose a quantum solution to the classical private information retrieval (PIR) problem, which allows one to query a database in a private manner. The protocol offers privacy thresholds and allows the user to obtain information from a database in a way that offers the potential adversary, in this model the database owner, no possibility of deterministically establishing the query contents. This protocol may also be viewed as a solution to the symmetrically private information retrieval problem in that it can offer database security (inability for a querying user to steal its contents). Compared to classical solutions, the protocol offers substantial improvement in terms of communication complexity. In comparison with the recent quantum private queries [Phys. Rev. Lett. 100, 230502 (2008)] protocol, it is more efficient in terms of communication complexity and the number of rounds, while offering a clear privacy parameter. We discuss the security of the protocol and analyze its strengths and conclude that using this technique makes it challenging to obtain the unconditional (in the information-theoretic sense) privacy degree; nevertheless, in addition to being simple, the protocol still offers a privacy level. The oracle used in the protocol is inspired both by the classical computational PIR solutions as well as the Deutsch-Jozsa oracle.