Intel working on rootkit detection techniques
By
Tom Krazit
,
IDG News Service
, 12/07/2005
- Share/Email
- Tweet This
- Print
Intel is working on a research project that would immediately notify PC users if they inadvertently download a rootkit like the XCP (extended copy protection) software found on certain music CDs shipped by Sony , researchers said Tuesday.
Intel held an open house for press, analysts, students and employees in Folsom, Calif., Tuesday to showcase some of its projects
and talk a little about its vision of the future of computing. That future involves relieving humans of serving as the gatekeepers
for reams of information flowing between computers and people, said David Tennenhouse, vice president of Intel's Technology
Group and director of research at the company.
"We need to connect the computers directly to the data, so the human beings don't have to be the I/O channel, and elevate
the role of the human being to a more supervisory role," Tennenhouse said.
One interesting project involves placing a small chip on a PC's motherboard to constantly monitor programs for modifications
that might be the result of a malicious attack, said Travis Schluessler, a researcher with Intel.
Sony's XCP software implemented copy-protection policies with rootkit software. Rootkits are pieces of software designed to
access a system and make changes or implement policies without being detected by the operating system or antivirus software.
Security experts say malicious hackers might have used Sony's rootkit software to launch undetectable attacks.
Security vendors recently admitted that Sony's XCP rootkit caught them by surprise, even though it had been installed on thousands
of systems for months before an independent researcher identified it, and their products need significant upgrades to detect
rootkits.
The idea behind the Intel project is to protect systems from malicious programs that make their way onto a system and attack
application software running in the system's memory, Schluessler said. Many modern worms and viruses, such as the Slammer
and Blaster worms, attempt to disable programs running in memory or alter those programs to run the attacker's code and then
propagate themselves across a network, he said.
The succinctly named "OS Independent Run-Time System Integrity Services" project attempts to limit memory-resident attacks
by detecting changes in application code as they happen, allowing IT administrators to take immediate action, Schluessler
said. Under this scenario, an "integrity measurement manager" running on a chip outside of the main CPU (central processing
unit) or memory would identify a rootkit or malware that started to make changes to the program in memory. That detection
would trigger any number of responses set by the IT department.
For example, an infected PC could be set to immediately detach from the network when an alert is triggered, preventing the
worm or attack from spreading beyond that PC, Schluessler said. The alert could also send an e-mail or pop-up message to the
network administrator informing them of the intrusion.
Intel doesn't expect its project to take the place of anti-virus or anti-spyware software, but believes it could supplement
them, Schluessler said. Malware often attempts to shut down or alter antivirus software to make way for future attacks, and
this project could back up the antivirus software, or "check the checker," he said.
The IDG News Service is a Network World affiliate.
Comment