"A Faster Way to Preserve Privacy Online"
Searching the Internet can expose information that a user would prefer to keep private. For example, when someone searches for medical symptoms online, they may be disclosing their health conditions to Google, an online medical database such as WebMD, and possibly hundreds of these companies' advertisers and business partners. Researchers have long been developing techniques that allow users to search for and retrieve information from a database privately. However, these methods are still too slow to be used effectively. Therefore, MIT researchers developed a method for retrieving private information approximately 30 times faster than comparable methods. Their method allows users to search an online database without disclosing their query to the server. Furthermore, it is driven by a simple algorithm that would be easier to implement than previous work's more complicated approaches. Their method could allow for private communication by preventing a messaging app from knowing what users are saying or who they are communicating with. It could also be used to retrieve relevant online advertisements without advertising servers learning about a user's preferences. The MIT researchers devised Simple PIR, a protocol in which the server performs much of the underlying cryptographic work before a client sends a query. This preprocessing step generates a data structure that contains compressed information about the database contents and is downloaded by the client before sending a query. The data structure serves as a hint to the client about what is contained in the database. Once the client has this hint, it can make an infinite number of queries, and these queries will be much smaller in terms of both the size of the messages sent by the user and the work that the server needs to do. To reduce the size of the hint, the researchers created a second technique called Double PIR, which essentially involves running the Simple PIR scheme twice. This article continues to discuss the fast method developed by MIT researchers to enable users to search for information without revealing their queries.