While YouTube (as well as other online content portals) has taken steps to protect viewers’ privacy, it isn’t enough and intelligence agencies, hackers and online advertising companies can still determine the viewing habits of its users.
The report comes from Israel-based Ben-Gurion University of the Negev (BGU) that presented the research at the recent Black Hat Europe cybersecurity meeting in London.
Researchers developed a machine-learning algorithm that identifies which video YouTube viewers watched with a high degree of accuracy. The algorithm was based on how video services work, how video content is encoded and how a video player requests information to play it.
The algorithm was used to determine if a YouTube viewer watched a specific video from a set of suspicious, terror-related videos. Researchers believe using a similar algorithm could allow intelligence agencies the ability to track terrorists or other suspicious individuals or allow marketing companies to track the number and make-up of viewers watching a particular ad from a video.
The information could be helpful, but researchers warn YouTube users to be aware that their viewing history on YouTube or other internet video platforms can be tracked.
“It's important to know that video encryption is not as secure as we once thought,” says Ran Dubin, a doctoral student in BGU’s Department of Communication Systems Engineering. “Google, YouTube's parent company, is not likely to patch the gaps, since it would be prohibitively expensive to create a traffic obfuscation mechanism for every user's every video request.”
However, an upside to this research is the ability to assess video quality without breaking an encryption. Internet service providers want to ensure high-quality streaming but encryption has made accessing this information much more challenging. “This tracking algorithm could help,” Dubin says.