Web Isolation Technology


BACKGROUND: The vast majority of malicious code on the web is on legitimate pages that have been hacked for the purpose of infecting anyone who views the hacked web resource. High use sites that are infected will cast a wide net for unsuspecting users, but often a particular high value target or a segment of the population is being targeted. To do this hackers will infect sites or advertisements that are displayed to the targeted users. Privacy is also almost non-existent on the web. Search providers and data aggregators will know what everybody in an organization is working on and can also discern organization wide policies by viewing spikes in traffic originating from that organization.

SOLUTION: Separate the functions that the browser performs into two different platforms. The client browser will execute only the benign code required for display of the web page (HTML, CSS). The “Rendering” browser will process the complete code for the page and transfer the document object model updates to the client browser. When trackers want to follow what you are doing they will only see the rendering browser, not the client. Also since the rendering browser is in a Virtual Machine it will be refreshed regularly, erasing all record of the browsing session.

WHY CURRENT SECURITY METHODS FAIL: Signature Analysis uses known signatures of malicious code to block malware but it only detects and disables malware that was already known to be in existence. Blacklisting of URLs has the same failing as signature analysis. Script blocking is able to block all threats but it makes many pages unusable.

WHY CURRENT PRIVACY METHODS FAIL: Blocking header information and IP address is like hiding the license plate on a pink Cadillac. Fingerprinting methods pass through the TOR network uninterrupted. “Private browsing” does not hide who you are from the sites you visit. ”Do Not Track” requires voluntary participation by web sites.