Stories
Slash Boxes
Comments

SoylentNews is people

SoylentNews is powered by your submissions, so send in your scoop. Only 17 submissions in the queue.

Submission Preview

Link to Story

Machine Learning Sniffs Encrypted Web Traffic with High Accuracy

Accepted submission by JohnnyComputer mailto:jacob.ian.lee@gmail.com at 2014-06-21 13:06:10
Security

JohnnyComputer writes:

According to a new paper [arxiv.org], statistical analysis of encrypted web traffic can reveal far more than you suspect, including details about your instant chat messages such as what language you are writing in, and details about your browsing including which web sites you visit. iMessages, Whatsapp, and Viber were particularly vulnerable. Many of the attacks had impressive accuracy rates (80-100%). Read more here [technologyreview.com].

Original Submission