A recent Microsoft presentation on the ethics of disclosing vulnerabilities before a patch is available sparked a debate Thursday among experts who tended to lean in favor of releasing information sooner than later.
The joint presentation from Microsoft and Lancope, given this week at the Virus Bulletin conference in Berlin, brought a new twist to what has been a hot topic for years in the security industry.
Holly Stewart, senior program manager lead at Microsoft's Malware Protection Center, and Tim Cross, director of security research at Lancope, looked at pre-patch releases of vulnerabilities that were being exploited by cybercriminals. In looking at a number of cases over the last few years, the duo came to the conclusion that such releases caused a "significant bandwagoning effect" among attackers.
"When real attack activity confirms the practical value of a vulnerability, and attackers know that no one can defend themselves because patches aren't available, that's an opportunity that they tend to jump on in large numbers," Cross said on Lancope's blog.
One such case was the Stuxnet malware widely believed to have been developed by the U.S. and Israel to damage Iranian nuclear facilities. The malicious program targeted a previously unknown vulnerability in Windows.
Once Stuxnet was discovered and the vulnerability disclosed, the rate at which the latter was exploited by hackers exceeded the infection rate of the original Stuxnet exploit, Cross said.
Therefore, Microsoft and Lancope drew the conclusion that even when attacks are underway, synchronizing the disclosure of vulnerabilities with a software vendor's patch release made sense.
"Getting the right answer involves considering how quickly the vendor will be able to produce a patch, whether there are practical workarounds available before the patch comes out, and how quickly attack activity is spreading on the Internet," Cross said.
In polling a half dozen security experts, CSOonline found differences in degrees. While some believed releasing vulnerabilities should be the priority, others felt researchers should strive for cooperation with the software vendor.
Jeremiah Grossman, chief technology officer for WhiteHat Security, fell into the former camp.
"As a researcher, you want to give potential victims who want to protect themselves the information they need to do so," he said. "This means disclosing the vulnerability details when they become known.
"Those who are more complacent, who wouldn't put the vulnerability details to use anyway, are going to get hacked one way or the other."
Paul Henry, security and forensic analyst at Lumension, was more willing to compromise with vendors.
"I am a firm believer in responsible disclosure whereby the researcher and vendor agree on a time frame for the vendor to correct the issue before the researcher releases details on the issue," he said. "In my opinion releasing details prior to a vendor providing a patch does a dis-service to the community at large."
Wolfgang Kandek, chief technology officer for Qualys, also believed in working closely with the vendor with one caveat.
"If the vendor is unavailable or not willing to cooperate, the researcher should publish the vulnerability and, if possible, provide workarounds," he said.
Ron Gula, chief executive officer of Tenable Network Security, said researchers often have to follow their conscience on when to disclose a flaw, because of the amount of time tech vendors can take to distribute a patch.
While some are quick, others, such as makers of industrial control systems, can take years.
"We have found zero-day vulnerabilities at Tenable, just through the course of normal vulnerability testing, and it has taken the vendor two years to come up with a patch," Gula said.
Ultimately, responsible disclosure is subjective, since everyone has their own criteria for determining when the number of attacks against an undisclosed vulnerability is sufficient to warrant going public, Zak Dehlawi, senior security engineer for Security Innovation, said. Other judgment calls include whether the vendor is being cooperative.
"I view that if an attack is happening in the wild, then it's appropriate for security researchers to release details about the vulnerability," he said.
By doing so, pressure is placed on the software maker and anti-virus vendors to offer defenses, and system administrators can take steps to secure corporate computers.
Matthew Neely, director of research and development at SecureState, agreed. "I'm a believer that the more information that is available, the better for the defenders to be able to defend their systems."