The latest draft of the Cyber Security Framework (CSF) mandated by President Barack Obama in February fails to provide an effective battle plan for defending the nation's critical infrastructure, experts say.
In favoring a self-regulatory approach with industry, the proposed CSF provides no specific methodologies for securing the industrial control systems (ICS) found in water treatment facilities, power and manufacturing plants and energy pipelines, critics say.
The National Institute of Standards and Technology (NIST), which Obama put in charge of working with industry in formulating the CSF, published the draft last week.
On Wednesday, Ralph Langner, a renowned Hamburg, Germany-based consultant on ICS security, said the application of the CSF as it is written would provide no "measurable cybersecurity assurance."
"A fundamental problem of the CSF is that it is not a method that, if applied properly, would lead to predictable results," Langner said on a blog post.
Joe Weiss, a leading expert in ICS security and managing partner of Applied Control Solutions, agreed with Langner, saying the CSF draft fails to provide specific ways to achieve security. Instead, NIST is taking a useless "high-level approach" that favors self-regulation over mandates, he said.
"A high-level framework doesn't do us a hell of a lot of good," he said. "It's not useful at all for someone trying to secure an industrial control system -- period."
NIST issued a statement emphasizing that the current document is only a draft and more work is needed.
"The goal of the framework is to bring together existing standards, policies and best practices into a tool organizations can use to ensure effective cybersecurity," NIST said. "We put out an early discussion draft to ensure that we got feedback -- including positive and negative -- on the framework as currently presented. We hope to get active participation from all stakeholders."
[Also see: Insecure ICS, hacker trends prompt federal warnings]
In meeting with NIST officials at the RSA conference in February, Weiss drew the conclusion that the organization did not have a "good comprehensive understanding" of ICSs. This lack of knowledge is reflected in the CSF's approach toward determining the level of cyberattack risk faced by an ICS, he said.
Rather than analyze risk in terms of cyberthreats, NIST is using the traditional approach of determining risk based on frequency and severity of adverse consequences, Weiss said. This formula does not apply because it fails to take into account the dramatic rise in cyberattacks against critical infrastructure, only the fact that none have been successful.
"Traditional risk assessment doesn't address malicious intent," Weiss said. "It's that simple."
Langner made a similar point in his blog post, saying, "Regardless of the popularity of risk parlance, risk-based approaches in ICS security lack empirical foundation, and the outcome of a risk assessment can be stretched in any direction."'
Langner and Weiss also agreed on the ineffectiveness of the CSF proposal letting organizations choose the level of cybersecurity they want to achieve. Allowing a critical infrastructure provider to set its own goals means an organization could choose a level of zero, and "still be conformant with the CSF," Langner said.
"The CSF allows any organization, no matter how good or bad at cybersecurity, to be CSF-conformant," he said. "It makes everybody happy. Everybody, including potential attackers."
NIST is expected to publish the final CSF in February.