July 7, 2015 By Douglas Bonderud 2 min read

No one likes passwords. Users complain they’re too cumbersome, easily forgettable and make logging into multiple endpoints a pain. IT security teams, meanwhile, struggle against user predilection to choose easily guessed, familiar password strings and then daisy-chain them across multiple accounts.

According to Naked Security, however, credit card giant MasterCard is rolling out a solution for online purchases over the next few months: pay-by-face, which would require a facial scan as confirmation for purchases. There’s huge interest here and big-time potential, but do selfies really fit the picture of perfect security?

Blink and You’ll Buy It

According to MasterCard’s Chief Product Security Officer Ajay Bhalla, “Passwords are a pain.” He argued that the potential security offered by passwords is consistently undermined by users: While they want personal data protected online, they’re not happy about memorizing multiple sets of login credentials or following complicated password rules. That’s the impetus for the company’s new facial scanning program for online purchases.

Here’s the idea: Using the MasterCard phone app, consumers will be able to pay for their purchases online. Before the transaction is complete, however, the app will ask for user approval in the form of either a fingerprint or facial scan. For fingerprints, users need only touch their screen. For facial recognition, they need to stare at the screen for a moment and then blink once. Bhalla believes that the “new generation” of buyers will embrace this selfie technology, and he might be right. On paper, this sounds like the ultimate form of multifactor authentication; users can’t exactly leave their face at home or forget how to blink.

A Password About-Face?

Not surprisingly, MasterCard’s new master plan has raised a number of security concerns. The first stems from worry over data collection: Does the app allow MasterCard to amass a huge library of consumer selfies and leverage them for unknown future purposes? As noted by The Hacker News, however, the PCI giant says that all facial scans will be digitally converted and then sent to their secure servers, leaving them with no way to reconstruct user profiles.

The second concern revolves around malicious use. Technology companies have been trying for years to make facial recognition an essential part of mobile device security, but so far, even the most advanced systems have proven easy to break. Static face scans, for example, were easily fooled by photographs, and Google’s 2012 initiative “Liveness Check” was tricked by using two photoshopped images that cycled between a user’s face with eyes closed and eyes open, DroidDog reported.

The FBI, meanwhile, has quietly been leveraging the Next Generation Identification (NGI) system developed by Lockheed Martin, which contains over 25 million mugshots. While MasterCard said there’s no way to reconstruct facial data and everything is securely stored on its servers, it has been quiet about any encryption technology. If malicious actors were able to steal this data and reverse the digital conversion procedure or trick the system with altered images, the results could be disastrous. Unlike passwords, there’s no way for users to reset their face.

Is pay-by-face perfectly posed take the purchasing world by storm? There’s a good chance users will appreciate the ease of taking a selfie instead of remembering complicated passwords. Long-term value, however, depends on the ability of MasterCard’s detection software to perform near-flawlessly and avoid the stigma of being two-faced when it comes to the storage of consumers’ personal data.

More from

Unified endpoint management for purpose-based devices

4 min read - As purpose-built devices become increasingly common, the challenges associated with their unique management and security needs are becoming clear. What are purpose-built devices? Most fall under the category of rugged IoT devices typically used outside of an office environment and which often run on a different operating system than typical office devices. Examples include ruggedized tablets and smartphones, handheld scanners and kiosks. Many different industries are utilizing purpose-built devices, including travel and transportation, retail, warehouse and distribution, manufacturing (including automotive)…

Stealthy WailingCrab Malware misuses MQTT Messaging Protocol

14 min read - This article was made possible thanks to the hard work of writer Charlotte Hammond and contributions from Ole Villadsen and Kat Metrick. IBM X-Force researchers have been tracking developments to the WailingCrab malware family, in particular, those relating to its C2 communication mechanisms, which include misusing the Internet-of-Things (IoT) messaging protocol MQTT. WailingCrab, also known as WikiLoader, is a sophisticated, multi-component malware delivered almost exclusively by an initial access broker that X-Force tracks as Hive0133, which overlaps with TA544. WailingCrab…

Operationalize cyber risk quantification for smart security

4 min read - Organizations constantly face new tactics from cyber criminals who aim to compromise their most valuable assets. Yet despite evolving techniques, many security leaders still rely on subjective terms, such as low, medium and high, to communicate and manage cyber risk. These vague terms do not convey the necessary detail or insight to produce actionable outcomes that accurately identify, measure, manage and communicate cyber risks. As a result, executives and board members remain uninformed and ill-prepared to manage organizational risk effectively.…

Pentesting vs. Pentesting as a Service: Which is better?

5 min read - In today's quickly evolving cybersecurity landscape, organizations constantly seek the most effective ways to secure their digital assets. Penetration testing (pentesting) has emerged as a leading solution for identifying potential system vulnerabilities while closing security gaps that can lead to an attack. At the same time, a newer entrant into the security arena is Pentesting as a Service (PTaaS). Although PTaaS shares some similarities with pentesting, distinct differences make them two separate solutions. This article will discuss how these methodologies…

Topic updates

Get email updates and stay ahead of the latest threats to the security landscape, thought leadership and research.
Subscribe today