Computer crime, cyber crime, cyber terrorism problemsLogo - Computer Crime Research Center (CCRC)

Security Nightmare

DAN VERTON A new book by Notra Trulock, former director of intelligence at the U.S. Department of Energy (DOE), details how apathy and politics can infect the decision-making of senior managers and create a security nightmare for frontline administrators.

Code Name Kindred Spirit (Encounter Books, 2003) takes readers inside the Chinese nuclear espionage scandal that occurred between 1995 and 2001 at Los Alamos National Laboratory and led to the DOE's investigation of nuclear scientist Wen Ho Lee. Trulock also sheds new light on how senior DOE management contributed to the appalling state of computer and enterprise security that contributed to the possible disclosure of the nation's most sensitive nuclear secrets.

In an interview with Computerworld, Trulock offers insight into how mismanagement of security issues at the highest levels of an organization can have devastating results.

In your book, you wrote about many management and leadership shortcomings at the DOE, and especially how those shortcomings had a debilitating effect on enterprise security. Can you expand on a few examples that might offer some sober lessons for large corporations? Most importantly, I think, security was sharply devalued by the Energy Department and senior lab managers during the 1990s. It is common knowledge that budgets were run down and both manpower and training were sharply cut back. By far the worst offense, however, was the corruption and manipulation of security standards and measures of performance imposed by these officials.

Time and again, senior managers fudged or simply suppressed the results of security exercises, evaluations and self-assessments. More than once, senior lab managers were caught red-handed pressuring security officials to revise upward poor marks on security exercises. In other cases, the outcomes of such exercises were rigged to conceal vulnerabilities. Inevitably, senior management officials also had to suppress and silence those security officers unwilling to go along with the coverups.

You also wrote that you "witnessed many incidents in which CI [counterintelligence] officers were asked to report on the same people who wrote their performance appraisals, granted raises and approved budgets." What lessons and suggestions should come out of this for large corporate enterprise security programs that might be wrestling with finding the right management structure? I don't know whether the problems I witnessed are unique to the national labs. As counterintelligence was de-emphasized [along with security] at the labs in the 1990s, CI units were pushed further down the lab structure to the point that multiple layers of management existed between CI officers and the lab director. Consequently, CI [and security] problems rarely were brought to the attention of the lab directors or their senior management group. For a time at Los Alamos, the CI [group] reported to a unit that was also responsible for hallway maintenance and replacing burnt-out light bulbs.

First and foremost, large corporate enterprise security units should enjoy high visibility within the enterprise and report directly to either the chief executive or chief operating officers. The further down a [security] unit is in the corporate structure, the greater the potential for interference and obstruction in the performance of the unit's mission.

For companies that have large volumes of intellectual capital and proprietary research to protect, what is your advice on how to achieve an appropriate level of security while at the same time maintaining enough flexibility to remain competitive? The 1990s were a period in which senior managers refused to think through the long-term implications of losing secrets vs. the short-term political benefits of openness or declassification. In an enterprise setting, I suppose this would entail deciding what is worthy of protection and what senior managers might be willing to concede to potential competitors. The costs of lost or stolen data would have to be factored into such an equation. Flexibility might not be valued so highly if the loss of sensitive data undercut an enterprise's competitiveness.

You mention one instance in which a lab director simply crossed out classification markings on a document and released it to the media. We now know that the document in question contained sensitive information. What should corporate security managers do when confronted with similar situations? I assume you mean beyond updating his or her resume and looking for a new job? Security managers can try a slow, steady process of educating the CEO to the potential downsides of lax security. I know of one Fortune 100 corporation whose world-famous CEO totally disdained security of any type and made nearly all corporate locations open campuses. He retired just before 9/11 and left his successors a staggering problem of how to protect corporate assets worldwide against terrorists determined to attack symbols of American economic power. His successors had to start from scratch to rebuild security at the corporation's facilities.

You wrote about how an unprotected computer network at Los Alamos was "repeatedly attacked by computer hackers, several of them known to be working for foreign governments." Can you describe the DOE's response? A top-secret internal Justice Department review of the Los Alamos nuclear espionage case, eventually declassified in 2001, revealed that from 1988 to 1997, Wen Ho Lee moved nearly 1,600 files onto an unprotected Los Alamos network and portable magnetic tapes. These files included about 800MB of classified data.

Justice officials claimed, however, that while diagnostic tools available at Los Alamos enabled systems administrators to detect and log unauthorized access attempts, these tools could not identify exactly which files were being accessed. The diagnostic work that was done showed that on six different occasions after Lee had transferred files to the unprotected server, another visitor would log on using Lee's user ID and password. His defense team would later claim that this activity was simply his daughter logging on from college in California to play computer games on the Los Alamos server.

But a classified document listed 792 computer security incidents from October 1997 to June 1998 alone. The document identifies 324 of these as "attacks from systems outside the United States." The documents concluded that all such attacks sought to gain password files and many represented "actual compromises of DOE computer systems where the intruders gained access." Such access included "successful root compromises."

You write about the threat to U.S. businesses, particularly computer companies, from foreign competitors and even foreign government espionage. Specifically, you say that "commercial diplomacy" may have jeopardized national security by stripping the U.S. of its technological edge and allowing sensitive computer and other technologies to be exported. Given the current global crisis we are facing with "rogue" states possibly going nuclear, to what extent do you think the government's easing of export controls on advanced computer technology has threatened our overall security? There is clearly a line of demarcation between exports that do no harm to our national security and those that should be considered a clear and present danger. Along with many others, I believe it is America's scientific and technical potential that has provided the edge that has safeguarded our national security and even won the Cold War. Investing wisely enabled us to sustain that edge over any potential adversaries, although many worried during the 1990s that as a country, we were no longer willing to pay to sustain that advantage.

But beyond that, I can't help but note that export control relaxations that created a firestorm of controversy during the Clinton administration go almost unnoticed on President Bush's watch. In 2000, Energy Department scientists concluded that the availability of high-performance computers at speeds of 10,000 [millions of theoretical operations per second] would "be of significant use to China's [nuclear warhead] designers in examining likely gaps in their nuclear weapons programs." We know that China is a primary supplier of nuclear expertise and technologies to Pakistan. Pakistan, in turn, has been instrumental in assisting North Korea to develop its own nuclear program. Put simply, if Clinton administration decisions to raise export control thresholds represented a serious threat to U.S. national security, then what do Bush administration decisions to double Clinton's highest export benchmark represent?

Home | What's New | Articles | Links
Library | Staff | Contact Us

Copyright Computer Crime Research Center 2001, 2002 All Rights Reserved.
Contact the CCRC Office at +38 061 220 12 83