"Building a Better 'Canary Trap'"

A canary trap in espionage is the spread of multiple versions of false documents to hide a secret. The canary trap technique can be used to detect information leaks or create distractions that conceal valuable information. A team of researchers at Dartmouth’s Department of Computer Science developed a new data protection system called WE-FORGE that uses Artificial Intelligence (AI) to expand upon the canary trap method. The system protects intellectual property such as drug designs and military technologies by producing false documents. WE-FORGE improves upon the canary technique by using natural language processing to automatically generate multiple fake files that are sufficiently similar to the original ones to be believable but different enough to be incorrect. The system also adds randomness to prevent adversaries from identifying real documents. WE-FORGE can create many fake versions of any technical design document, thus making it significantly difficult for adversaries to determine which document is real once they have successfully hacked a system. The use of this technique causes adversaries to waste their time and resources, as well as have lower confidence. This article continues to discuss the concept of canary traps in espionage and how the WE-FORGE data protection system builds on this technique to better deceive would-be attackers.

Homeland Security News Wire reports "Building a Better 'Canary Trap'"