Researchers reveal how easily Signal, Telegram and WhatsApp messages can be hijacked

News by Davey Winder

Session-hijacking side-channel attacks can risk exposing users messages in full, researchers at Cisco Talos Intelligence Group have found.

New research reveals an old problem: secure messaging apps are only as secure as the system they are running on.

Organisations of all sizes have come to depend upon secure messaging apps such as Signal, Telegram and WhatsApp on the assumption that sensitive content shared this way will remain confidential, an assumption blown apart by new research from the Cisco Talos Intelligence Group.

Talos researchers have demonstrated how session-hijacking side-channel attacks put secure messaging applications including Signal, Telegram and WhatsApp, at risk of exposing user messages in full. Vitor Venture, the technical lead at Talos, states, "If an attacker can copy the session tokens from a desktop user, then it will be able to hijack the session. The attacker won't need anything else other than the information that is stored locally."

Importantly, it matters not one jot whether that information is actually encrypted; copying the data enables the threat actor to create a 'shadow session' and that is game over.

Although the specific methodologies and consequences vary between the messaging apps mentioned, the cryptographic protocols used by them (MT Protocol or Signal Protocol) for end-to-end encryption cannot protect user messages from attack vectors such as the UI framework.

Take the Electron framework that's used by both Signal and WhatsApp which has a known vulnerability that can lead to remote code execution or message copying. Throw in the fact that these apps also run on multiple platforms, including desktop machines, and security researchers can sniff out trouble.

If security researchers can smell that insecurity stink, you better bet that criminal noses are also twitching. Earlier this year, Talos published an analysis of the TeleGrab malware which hijacked sessions from Telegram.

The smell is now so bad that Talos researchers thought they'd apply the same technique used by TeleGrab against the messaging apps, sadly with great success.

So, where does this leave business users of so-called secure messaging apps? Should they continue to use them, indeed should they be using them to share sensitive information in the first place?

Wicus Ross, lead security researcher at SecureData Labs, reminds us that labs-based exploits don't necessarily translate into real-world attacks. "Most side-channel attacks are inefficient, expensive and complex compared to other more conventional means," Ross says. "There are easier ways to extract information from people or devices than trying to leak random data from a device."

Ron Masas, a security researcher at Imperva, isn't so sure. "Side-channel attacks on non-cryptosystems are probably the most overlooked form of attack we see today," he told SC Media UK. "The attack surface is huge and when combined with the low awareness, you have a real problem on your hands."

That said, Masas does point out that "if your operating system is compromised, there is little any app running on the system could do to stop the attacker from gaining sensitive data."

John Safa, founder of secure content delivery vendor Pushfor, is in no doubt that the answer to the second part of that question is a big no. Safa explained that using secure consumer messaging apps in business is a very risky.

"Most secure messaging apps have data held on servers that are not in the users’ control," Safa points out. "Content shared on secure messaging platforms are not typically designed to share content securely thus leaving it to be downloaded to the device, screenshot or copied."

That's without taking into account the small problem of ex-employees taking that content with them and the "zero accountability of shared data over a consumer messaging app," as Safa puts it, referring to there being no record of if, when and how content is shared on these networks.

"Just because it’s called a secure messaging app doesn’t mean the system that it’s bolted onto is secure," warns Daniel Smith, researcher for Radware’s Emergency Response Team. "Users need to understand you can’t download security. They need to be educated and understand that apps are not the end-all solution."

What should secure messaging platforms be doing to mitigate these vulnerabilities? Is there too much focus on the cryptographic protocols protecting data in transit (and at rest on their own servers) at the expense of protecting application state and user information by delegating this security process to the operating system instead?

"End-to-end encryption has raised the security bar considerably – securing data at rest is also an important function when looking at complete security," Ed Williams, EMEA director of SpiderLabs at Trustwave, says. "A third part of that has to be the user and they now need to bear some of the security burden."

As Williams says, "The ability to use a desktop to view messages is purely for convenience, and we know that security and convenience don’t always go hand-in-hand..."

Find this article useful?

Get more great articles like this in your inbox every lunchtime

Video and interviews