The Future Of Passwords

In a search to address security concerns and solve the problems associated with password protection, DARPA, (Defense Advanced Research Projects Agency) is now in the process of developing a behavior based biometric identification system that may replace traditional passwords in the future. According to DARPA, even the strongest password is still one of the weakest forms of security. Entering a password doesn’t prove who you are, it just proves that you know the correct combination to gain access.

The US military has been plagued with instances where individuals were able to gain access to private information via hacking or guessing the password. For this reason, DARPA has recruited the help from one of the most prestigious military institutions in the U.S. A multimillion dollar grant has been awarded to the West Point Academy. The grant comes from DARPA’s “Active Authentication” program, which seeks to replace traditional authentication techniques, such as passwords. The grant will be used by researchers to develop cognitive algorithms that will learn to recognize behavioral patterns based on how someone uses their devices. The idea is that the way each person uses their equipment is unique and that the program can determine who is using it by their behavior.

Also Read: I’ve Been Hacked – What To Do Next

How It Might Work

DARPA Vector Logo.epsIn working practice, the user would log in, and the device would monitor that users behavior to determine whether or not that person is who they appear to be. The user’s habits and behavior will be gathered and stored on a remote server. This profile will load every time the user logs on. If the user’s behavior deviates too much from what the program knows about the user, it will raise awareness to the system operator, or automatically shut down the device.

In a statement by DARPA,

The Active Authentication program seeks to address this problem by developing novel ways of validating the identity of the person at the console that focus on the unique aspects of the individual through the use of software based biometrics. Biometrics are defined as the characteristics used to uniquely recognize humans based on one or more intrinsic physical or behavioral traits. This program focuses on the behavioral traits that can be observed through how we interact with the world. Just as when you touch something your finger you leave behind a fingerprint, when you interact with technology you do so in a pattern based on how your mind processes information, leaving behind a “cognitive fingerprint.”

An example of this would be if you were in a habit of signing into your computer and pulling up the latest news from Hacked.com, this new level of security would keep track of things such as your WPM, the use of Firefox over Chrome, the applications you normally run, etc. Any major deviation from your habits, such as someone typing much faster than you normally do, might set off an alarm.

Security Concerns

To keep this highly personal data secure, DARPA says that the device doesn’t send all the data stored on the main server. Instead, the program combs through the data collected, and rates it based on how close to the user it believes it to be. The device then sends that score to the administrator who can decide whether to continue to allow access, or shut the device down.

This would also prevent constant lockouts as a person’s behavior changes. By sending only the score of the user through the network, it leaves virtually nothing of value for a hacker to steal.

Richard Guidorizzi, who until recently was the director of the “Active Authentication” program, states that,

I’m not trying to create the next database to be hacked that has everybody’s biometric behavior in the world. We’re not even storing your personal information, all we’re doing is reading it and developing a profile score and saying, ‘OK, this is in the range.’

Featured image from Shutterstock.

Author:
A UNC Chapel Hill graduate, blockchain enthusiast and analyst. I have a background in programming and IT, strong studies in econ, stats and game theory. I'm interested in online privacy and privacy laws.