That was the message delivered by speakers and other experts at the annual AppSec USA conference here last week. The conference, which focuses on application security and secure software development, featured some of the best-known experts in the field. One of their common themes: Application security vulnerabilities can come up at any time, even after software is vetted and deployed.
"Look at the world of health care," said Jeff Williams, CEO of application security vendor Aspect Security, in a presentation at the conference. "It's no longer enough, in many cases, to wait for the patient to come in once a year for a checkup. They're equipping the body with sensors that can measure blood sugar or heart rate, and then send you a warning. In some of those cases, your phone knows you're sick before you do.
"Application security needs to follow that same model. The once-a-year scan for vulnerabilities isn't working. Application security needs to happen continuously, in real time, and not just on some apps, but on a portfolio scale."
Williams advocated the use of vulnerability sensors, which enterprises can develop themselves, to help detect and warn security professionals and developers of newly discovered vulnerabilities in software.
"You can develop sensors to detect clickjacking vulnerabilities or injection vulnerabilities or just about anything," Williams said. "You can get your developers to build sensors directly into the application that will warn you when a vulnerability occurs, in real time."
Bala Venkat, chief marketing officer at application security vendor Cenzic, agreed that vulnerability scanning should be a continuous process. "Most enterprises do careful scanning during the predeployment process, but they stop there," he noted. "Once the application is in operation, they look for vulnerabilities only rarely or not at all. And that's why so many applications today have vulnerabilities that haven't been remediated."
Venkat advocates an ongoing approach to vulnerability scanning that includes analysis not only before deployment, but while the software is operating. "Some IT organizations are afraid to do this because they are worried that scans might affect the performance of an operating application or cause a service interruption. But the risks of not remediating a known vulnerability generally are far greater."
Veracode, another application security vendor, has implemented an internal process for application monitoring that requires developers not only to do a one-time check for security vulnerabilities, but to continuously monitor for problems throughout the life of the application.
"If you want developers to learn something about security, you have to make sure that you are continuously exposing them to the security issue," said Chris Eng, vice president of research at Veracode, who also spoke at the conference. "Otherwise, it's like teaching them a math concept that they learn once and never use again. It has to be part of the process."
"The technology for finding vulnerabilities is a lot better than it was even a couple of years ago," noted Robert Hansen, director of product management and technology evangelist at WhiteHat Security. "What we need to do is update the process to reflect that better technology."
Have a comment on this story? Please click "Add a Comment" below. If you'd like to contact Dark Reading's editors directly, send us a message.