DHS to Require Federal Agencies to Set Vulnerability Disclosure Policies

The US government will require each civilian agency to create a public policy for software-vulnerability disclosure, as well as a strategy for handling any potential security weaknesses reported by researchers. 

In the statement posted online, the US Department of Homeland Security's Cybersecurity and Infrastructure Security Agency (CISA) raised concerns that most civilian agencies' lack of disclosure policies will lead to confusion, a lack of faith that issues are getting fixed, and the fear of potential legal action. The requirement of a vulnerability disclosure policy (VDP) will mean that every vulnerability reporter will know what to expect when they find and report a software flaw.

"A VDP allows people who have 'seen something' to 'say something' to those who can fix it," CISA said in its announcement. "It makes clear that an agency welcomes and authorizes good faith security research on specific, internet-accessible systems."

The move is the latest by the US government to work with security researchers and hackers to find — and plug — the weaknesses in its Internet-connected systems. In 2016, for example, the US Department of Defense announced its digital disclosure policy and launched the Hack the Pentagon challenge, the first-ever bug bounty for the federal government. Since then, every branch of the military has had a bug bounty challenge.

Without a vulnerability disclosure policy, good-faith researchers will not come forward with information on a security weakness outside of such programs, says Chris Wysopal, chief technology officer for software-security provider Veracode.

"[Having] no VDP is definitely a chilling effect of receiving good information from cooperative researchers," Wysopal says. "They don’t know if an organization will be friendly to researchers contacting with vulnerability information."

Game Change

The effort is a significant change in how government agencies treat vulnerability reporting. Not even a decade ago, the feds pursued charges against many vulnerability reporters. In 2005, for example, federal prosecutors charged information-technology specialist Eric McCarty with hacking after he used a Web-application vulnerability to access a handful of records on the application website of the University of California. McCarty reported the issue to the press.

Private industry also has targeted researchers. In 2017, one e-mail marketing firm pursued security research Chris Vickery, now at Upguard, for discovering that the company had left 1.4 billion email accounts users accounts in a publicly accessible cloud server — records that pointed to the company as a significant spam operation.

Federal agencies do not, in general, have a formal way of allowing security researchers or other third parties to submit information about vulnerabilities to them. And if an outside researcher or group did submit an issue, there is no process for handling that report and acting on it, the CISA stated in draft document, Binding Operational Directive 20-01, Develop and Publish a Vulnerability Disclosure Policy. 

"These circumstances create an environment that delays or discourages the public from reporting potential information security problems to the government, which can prevent these issues from being discovered and fixed before they are exploited or publicly disclosed," the agency said.

The Binding Operational Directive requires that civilian agencies create a security contact with 15 days and publish a vulnerability disclosure policy within 6 months. The BOD also mandates that agencies collect data on reported vulnerabilities and how quickly they are remediated. DHS is soliciting comments on the proposed rule until Dec. 27.

"If structured properly, VDPs can provide a secure channel for disclosures that do not prematurely inform adversaries of flaws in websites or applications," says Brian Fox, chief technology officer and co-founder of software-management firm Sonatype. "Clear documentation would be needed by the VDP program on what is allowed in regards to finding versus exploiting a vulnerability. Paired with a strong communication protocol, researchers and VDP program teams can safely work through remediation efforts without political or national security implications."

While the proposed rule is a good first step, do not expect to see a rise in attempts to find vulnerabilities in production sites, says Veracode's Wysopal. 

"I don't think researchers are going to test government sites without an explicit VDP," he says. "There is perceived risk that the government will be more sensitive to having security issues exposed than corporations are."

A public comment can be submitted through GitHub.

Related Content

Check out The Edge, Dark Reading's new section for features, threat data, and in-depth perspectives. Today's top story: "A Cause You Care About Needs Your Cybersecurity Help."