Academia is bypassing the slowness with which mobile security vendors share information in a highly competitive market by launching a collaborative effort to battle an upsurge of Android malware.
Vendors are welcome, but not a deal breaker in the Android Malware Genome Project launched last week (May 22) at the IEEE Symposium on Security Privacy in San Francisco. The initiative is the brainchild of Xuxian Jiang, an assistant professor in the Computer Science Department at North Carolina State University.
Jiang and his research team kicked off the effort by contributing their collection of more than 1,200 malware in 49 different families. The contribution, which was sent via the project to 40 universities, research labs and security vendors, covers most of the existing malware, Jiang said.
Since the debut of Android malware in August 2010, the amount has grown astronomically as the number of Android smartphones has soared to account for nearly 60 percent of the market in the first quarter, according to International Data Corp.
The phenomenal growth has brought many inexpensive smartphones to consumers, while also attracting cyber-criminals. The number of threats to Android devices quadrupled in the first quarter from a year ago, says Intel-owned McAfee.
[See more in our Mobile/Wireless Security topic center]
With so much hacker activity, Jiang believed a collaborative effort was needed among research labs in and out of universities to build defenses against the menace faster. While security vendors are working feverishly to develop their own products, competition often slows the sharing of research, which hampers development of the most effective technology.
The goal of the Genome Project is to build better cooperative competition among vendors, while also helping nonprofit researchers. “To be honest, I don’t think our current situation will immediately change, but no change will be made if we as a community don’t attempt to make these efforts,” Jiang told CSO on Friday in an email interview.
The lack of collaboration among vendors has had an impact on the effectiveness of anti-virus software aimed at Android malware. In his research paper ( PDF document) presented at the IEEE Symposium, Jiang found the amount of malware detected by AV software available today ranged from 20 percent to 80 percent. Performance could be improved with more sharing of malware samples, Jiang said.
The Genome Project will make research, malware samples and mappings available to any organization submitting a formal request that includes verification of who they are. Recipients will not be required to share data in return.
One of the recipients of Jiang’s malware collection was vendor Bit9. Chief Technology Officer Harry Sverdlove acknowledged that information sharing among mobile security companies is scarce, because the market is young and rivals are trying to establish as strong a foothold as possible.
“In the mobile space, it really is a greenfield,” Sverdlove said. “Whoever’s got the new techniques, the new heuristics, they’re not just one or two [anti-virus] signatures ahead of the competition, they could be significantly ahead.”
Nevertheless, the pressure to cooperate is mounting, as the U.S. government pushes the industry to co-operate on cyber-security to better protect critical information technology systems. “The threat landscape is changing so quickly that the best defense that we can do is to share information,” Sverdlove said.
Read more about wireless/mobile security in CSOonline’s Wireless/Mobile Security section.