It’s been easy, amid the accusations and counteraccusations, to lose sight of the underlying seriousness of the charges. If the hacking claims are true, we are looking at a truly dangerous crisis that puts America’s democratic system at risk.
The gravity of the allegation calls for a calm, measured, meticulously documented inquiry—pretty much the opposite of what we’ve seen so far. The level of wild assertion has gotten to the point that some of the most respected pro-Western voices in Russia’s opposition have expressed alarm. As much as they despise Putin, they don’t buy the bungled investigations. “In the real world outside of soap operas and spy novels . . . any conclusions concerning the hackers’ identity, motives and goals need to be based on solid, demonstrable evidence,” wrote Leonid Bershidsky. “At this point, it’s inadequate. This is particularly unfortunate given that the DNC hacks were among the defining events of the raging propaganda wars of 2016.”
The lack of credible evidence, the opaque nature of cyber attacks, the partisan squabbles and smears, and the national-security fearmongering have all made this particular scandal very difficult to navigate. It may be years before we find out what really happened.
Fancy Bears, Cozy Bears—Oh My!
Take CrowdStrike, the hottest cybersecurity firm operating today. Based in Irvine, California, CrowdStrike was
launched in 2012 by two veterans of the cyber-attribution business: George Kurtz and Dmitri Alperovitch. Both previously worked for McAfee, an antivirus-turned-massive-cybersecurity firm now partially owned by Intel. But Kurtz and Alperovitch saw a market opportunity for a new boutique type of cyber-defense outfit and decided to strike out on their own. They also brought on board Shawn Henry, a top FBI official who had been in charge of running the agency’s worldwide cyber investigations.
CrowdStrike positioned itself as a next-generation full-service cybersecurity firm. Company officials argued that cybersecurity was no longer just about defense—there was too much data and too many ways of getting at it to protect everything all the time. You had to know your attacker. “Knowing their capabilities, objectives, and the way they go about executing on them is the missing piece of the puzzle in today’s defensive security technologies,”
wrote CrowdStrike cofounder George Kurtz. “By identifying the adversary . . . we can hit them where it counts.”
CrowdStrike hit the big time in 2015 with a $100 million infusion from Google Capital (now Capital G), Google’s
first-ever investment in a cybersecurity company. It was good timing, because CrowdStrike was about to be catapulted into the front ranks of cyber-threat assessors. Sometime in April or May, CrowdStrike got a call from the Democratic National Committee to investigate a possible intrusion into their servers. The company’s investigators worked with surprising efficiency. As one DNC insider explained to the
New York Times, the company was able to make a definite attribution within a day. There was no doubt, CrowdStrike told its DNC clients—the Russian government did it.
Behold, virtuous American republic, the degradation Vladimir Putin has in store for you!
The results of CrowdStrike’s investigation were first broken by the
Washington Post and then followed up in greater detail by CrowdStrike itself. In a post entitled “
Bears in the Midst,” Dmitri Alperovitch attributed the hack to two distinct and very nefarious “Russian espionage” groups: Cozy Bear and Fancy Bear, among the most sophisticated cyber-operators CrowdStrike had ever come across. “In fact, our team considers them some of the best adversaries out of all the numerous nation-state, criminal and hacktivist/terrorist groups we encounter on a daily basis,” he wrote. “Their tradecraft is superb, operational security second to none and the extensive usage of ‘living-off-the-land’ techniques enables them to easily bypass many security solutions they encounter.”
These cyberspooks were allegedly behind a string of recent attacks on American corporations and think tanks, as well as recent penetrations of the unclassified networks of the State Department, the White House, and the U.S. Joint Chiefs of Staff. According to CrowdStrike, Cozy Bear was most likely the FSB, while Fancy Bear was linked to the “GRU, Russia’s premier military intelligence service.”
Here, the cyber experts were telling us, was conclusive evidence that both the FSB and the GRU targeted the central apparatus of the Democratic Party. CrowdStrike’s findings didn’t just cause a sensation; they carpet-bombed the news cycle. Reports that Vladimir Putin had tried to hack America’s democratic process raced around the world, making newspaper front pages and setting off nonstop cable news chatter.
The story got even hotter after a hacker who called himself Guccifer 2.0 suddenly appeared. He took credit for the DNC hack, called CrowdStrike’s investigation a fraud, and began leaking select documents pilfered from the DNC—including a spreadsheet containing names and addresses of the DNC’s biggest donors. The story finally started going nuclear when WikiLeaks somehow got hold of the entire DNC email archive and began dribbling the data out to the public.
A Terrible System
CrowdStrike stuck to its guns, and other cybersecurity firms and experts likewise clamored to confirm its findings: Russia was behind the attack.
Most journalists took these security savants at their word, not bothering to investigate or vet their forensic methods or look at the way CrowdStrike arrived at its conclusions. And how could they? They were the experts. If you couldn’t trust CrowdStrike and company, who could you trust?
Unfortunately,
there were big problems with CrowdStrike’s account. For one thing, the names of the two Russian espionage groups that CrowdStrike supposedly caught, Cozy Bear and Fancy Bear, were a fiction. Cozy Bear and Fancy Bear are what cyber monitors call “Advanced Persistent Threats,” or APTs. When investigators analyze an intrusion, they look at the tools and methods that the hackers used to get inside: source code, language settings, compiler times, time zones, IP settings, and so on. They then compare all these things against a database of previously recorded hacks that is shared among cyber professionals. If the attack fits an old profile, they assign it to an existing APT. If they find something new, they create a group and give it an official name (say, APT911) and then a cooler moniker they can throw around in their reports (say, TrumpDump).