Revelations by The New York Times that President Barack Obama in his role as commander in chief ordered the Stuxnet cyberattack against Iran's uranium-enrichment facility two years ago in cahoots with Israel is generating controversy, with Washington in an uproar over national-security leaks. But the important question is whether this covert action of sabotage against Iran, the first known major cyberattack authorized by a U.S. president, is the right course for the country to take. Are secret cyberattacks helping the U.S. solve geopolitical problems or actually making things worse?
Bruce Schneier, noted security expert and author, whose most recent book is "Liars and Outliers," argues the U.S. made a mistake with Stuxnet, and he discusses why it's important for the world to tackle cyber-arms control now in an interview with Network World senior editor Ellen Messmer.
SLIDESHOW: Worst data breaches of 2012 -- so far
The question is going to be debated whether Stuxnet was a good tactic to stop Iran from developing a nuclear weapon by sabotaging its facility through a malware attack in a covert action that was ultimately discovered. In an interview with Chris Wallace on Fox News last night, former National Security Agency director, retired Gen. Michael Hayden, said he thought it amounted to "taunting Iran." Based on the mix of military leadership, governmental leadership and ethical questions it raises, is Stuxnet a suitable approach?
There are two parts to this analysis. The first is tactical: Is a cyber-weapon more or less suitable than a conventional weapon? In 2007 Israel attacked a Syrian nuclear facility; it was a conventional attack with warplanes and bombs. Comparing the two, Stuxnet seems far more humane -- even though it damaged networks outside of Iran. The other part to the analysis is more strategic. Stuxnet didn't just damage the Natanz nuclear facility; it damaged the U.S.'s credibility as a fair arbiter and force for peace in cyberspace. Its effects will be felt as other countries ramp up their offensive cyberspace capabilities in response. For that reason, Stuxnet was a destabilizing and dangerous course of action.
David Sanger's NY Times article of June 1, headlined "Obama order sped up wave of cyberattacks against Iran," offers a vivid account of how President Obama decided cyberattacks against Iran should proceed through cooperation with Israel through use of the Stuxnet malware. However effective this might have been in stopping Iran from developing a nuclear weapon, it's now widely thought that the Stuxnet malware got out of control, spreading in the wild. What's your view on this, assuming the Times article is fully accurate?
It seems to be correct.
Sanger's article was very interesting, and it is worth reading, but it basically confirmed everything we all knew. We knew that Stuxnet was the work of Israel and the United States. We knew that it was intended as a pinpoint attack, and spread beyond its intended target. Other investigative journalists uncovered these truths already. What Sanger's article added to the discussion was detail about the program from inside both the Obama and the Bush administrations.
Richard Clarke's book "Cyber War" draws the distinction between cyber-espionage and cyberattacks. He argues cyber-espionage should basically be considered a routine, acceptable practice of any country as part of government intelligence operations. But he argues other state-sponsored operations, such as putting malware secretly into a power grid for example, or launching an actual attack, is distinctly different, and has to be considered in the realm of offensive weapons. Clarke suggests cyberweapons should be subject to arms control agreements of various sorts much as other types of weapons that can be used in war are today. Do you draw the distinction between cyber-espionage and cyberweapons along these lines? And should there be an effort by the U.S. and others to craft treaties related to cyber-arms?
Of course there's a difference between intelligence gathering and offensive military actions. Throughout history, there has been a bright line between the two. And what's true in the geopolitics of the physical world is no different in cyberspace. This same distinction also exists in computer security more generally. There is a fundamental difference between passive eavesdropping attacks and more active attacks that delete or overwrite data. As to arms control agreements, I think it is vital for both society and cyberspace that we begin these discussions now. We're in the early years of a cyberwar arms race, an arms race that will be expensive, destabilizing, and dangerously damaging. It will lead to the militarization of cyberspace, and the transformation of the Internet into something much less free and open. Perhaps it's too late to reverse this trend -- certainly you can argue that military grade cyberweapons like Stuxnet and Flame have already destroyed the U.S.'s credibility as a leader for a free and open Internet -- but the only chance we have are cyberweapons treaties.
If so, how do you think that should proceed?
I'm not an idealist. I know that cyberwar treaties will be difficult to negotiate and even more difficult to enforce. Given how easy it is for a country to hide a chemical weapons plant, I know that it will be even easier to hide a cyberweapons plant. I also know that there is a lot of money and power trying to sow cyberwar fears.
But even with all of this, I think there is enormous value in the treaty process -- and in the treaties themselves. I think we need to proceed by starting the dialogue. We made a mistake with Stuxnet: We traded a small short-term gain for a large longer-term loss. We can't undo that, but we can do better in the future.
Read more about wide area network in Network World's Wide Area Network section.