New research has revealed that 14 of the world’s most popular VPN services leak sensitive customer data – in many cases because they don’t yet protect IPv6 traffic.
Researchers from Queen Mary University of London (QMUL) and the University of Rome launched the findings in a paper at the Privacy Enhancing Technologies Symposium today.
They analyzed the software in a hostile test environment, connecting devices to a Wi-Fi access point which hackers could theoretically be scanning for activity.
The team tried two different approaches: ‘passive observer’ attacks where the hacker attempts to collect unencrypted info passing through an access point; and more active DNS hijacking attacks where the adversaries redirect traffic to a server they control.
Worryingly, they found that all but four of the 14 providers studied were vulnerable to so-called ‘IPv6 leakage’, while all but one were exposed to DNS hijacking.
Information leakage included not just website addresses but the content of communications.
The paper explained the following:
“Whereas our work initially started as a general exploration, we soon discovered that a serious vulnerability, IPv6 traffic leakage, is pervasive across nearly all VPN services. In many cases, we measured the entirety of a client’s IPv6 traffic being leaked over the native interface. A further security screening revealed two DNS hijacking attacks that allow us to gain access to all of a victim’s traffic. The most alarming situation is where individuals use VPN services to protect themselves from monitoring in oppressive regimes. In such cases, users who believe themselves to be anonymous and secure will be in fact fully exposing their data and online activity footprint. In countries with state-maintained network infrastructures, it is likely that such monitors could take on the role of any of our adversarial models.”
The group also tested VPNs on mobile platforms and found that while Apple users are safe from IPv6 leakage – as the protocol is disabled on the platform – it affects all VPN services on Android.
The researchers argued that misinformation in the market means many consumers are unable to “tell apart vague and bold claims typical of product advertisement campaigns with actual facts.”