WikiLeaks on Tuesday dropped one of its most explosive word bombs ever: A secret trove of documents apparently stolen from the U.S. Central Intelligence Agency (CIA) detailing methods of hacking everything from smart phones and TVs to compromising Internet routers and computers. KrebsOnSecurity is still digesting much of this fascinating data cache, but here are some first impressions based on what I’ve seen so far.
First, to quickly recap what happened: In a post on its site, WikiLeaks said the release — dubbed “Vault 7” — was the largest-ever publication of confidential documents on the agency. WikiLeaks is promising a series of these document caches; this first one includes more than 8,700 files allegedly taken from a high-security network inside CIA’s Center for Cyber Intelligence in Langley, Va.
“Recently, the CIA lost control of the majority of its hacking arsenal including malware, viruses, trojans, weaponized ‘zero day’ exploits, malware remote control systems and associated documentation,” WikiLeaks wrote. “This extraordinary collection, which amounts to more than several hundred million lines of code, gives its possessor the entire hacking capacity of the CIA. The archive appears to have been circulated among former U.S. government hackers and contractors in an unauthorized manner, one of whom has provided WikiLeaks with portions of the archive.”
Wikileaks said it was calling attention to the CIA’s global covert hacking program, its malware arsenal and dozens of weaponized exploits against “a wide range of U.S. and European company products, include Apple’s iPhone, Google’s Android and Microsoft’s Windows and even Samsung TVs, which are turned into covert microphones.”
The documents for the most part don’t appear to include the computer code needed to exploit previously unknown flaws in these products, although WikiLeaks says that stuff may show up in a future dump. This collection is probably best thought of as an internal corporate wiki used by multiple CIA researchers who methodically found and documented weaknesses in a variety of popular commercial and consumer electronics.
For example, the data dump lists a number of exploit “modules” available to compromise various models of consumer routers made by companies like Linksys, Microtik and Zyxel, to name a few. CIA researchers also collated several pages worth of probing and testing weaknesses in business-class devices from Cisco, whose powerful routers carry a decent portion of the Internet’s traffic on any given day. Craig Dods, a researcher with Cisco’s rival Juniper, delves into greater detail on the Cisco bugs for anyone interested (Dods says he found no exploits for Juniper products in the cache, yet).
WHILE MY SMART TV GENTLY WEEPS
Some of the exploits discussed in these leaked CIA documents appear to reference full-on, remote access vulnerabilities. However, a great many of the documents I’ve looked at seem to refer to attack concepts or half-finished exploits that demand more work and that may be limited by very specific requirements — such as physical access to the targeted device.
The “Weeping Angel” project’s page from 2014 is a prime example: It discusses ways to turn certain 2013-model Samsung “smart TVs” into remote listening devices; methods for disabling the LED lights that indicate the TV is on; and suggestions for fixing a problem with the exploit in which the WiFi interface on the TV is disabled when the exploit is run.
ToDo / Future Work:
Build a console cable
Turn on or leave WiFi turned on in Fake-Off mode
Parse unencrypted audio collection
Clean-up the file format of saved audio. Add encryption??
According to the documentation, Weeping Angel worked as long as the target hadn’t upgraded the firmware on the Samsung TVs, and that the firmware upgrade eliminated the “current installation method,” which apparently required the insertion of a booby-trapped USB device into the TV.
Don’t get me wrong: This is a serious leak of fairly sensitive information. And I sincerely hope Wikileaks decides to work with researchers and vendors to coordinate the patching of flaws leveraged by the as-yet unreleased exploit code archive that apparently accompanies this documentation from the CIA.
But in reading the media coverage of this leak, one might be led to believe that even if you are among the small minority of Americans who have chosen to migrate more of their communications to privacy-enhancing technologies like Signal or WhatsApp, it’s all futility because the CIA can break it anyway.
Perhaps a future cache of documents from this CIA division will change things on this front, but an admittedly cursory examination of these documents indicates that the CIA’s methods for weakening the privacy of these tools all seem to require attackers to first succeed in deeply subverting the security of the mobile device — either through a remote-access vulnerability in the underlying operating system or via physical access to the target’s phone.
As Bloomberg’s tech op-ed writer Leonid Bershidsky notes, the documentation released here shows that these attacks are “not about mass surveillance — something that should bother the vast majority of internet users — but about monitoring specific targets.”
By way of example, Bershidsky points to a tweet yesterday from Open Whisper Systems (the makers of the Signal private messaging app) which observes that, “The CIA/Wikileaks story today is about getting malware onto phones, none of the exploits are in Signal or break Signal Protocol encryption.”
The company went to say that because more online services are now using end-to-end encryption to prevent prying eyes from reading communications that are intercepted in-transit, intelligence agencies are being pushed “from undetectable mass surveillance to expensive, high-risk, targeted attacks.”
As limited as some of these exploits appear to be, the methodical approach of the countless CIA researchers who apparently collaborated to unearth these flaws is impressive and speaks to a key problem with most commercial hardware and software today: The vast majority of vendors would rather spend the time and money marketing their products than embark on the costly, frustrating, time-consuming and continuous process of stress-testing their own products and working with a range of researchers to find these types of vulnerabilities before the CIA or other nation-state-level hackers can.
Of course, not every company has a budget of hundreds of millions of dollars just to do basic security research. According to this NBC News report from October 2016, the CIA’s Center for Cyber Intelligence (the alleged source of the documents discussed in this story) has a staff of hundreds and a budget in the hundreds of millions: Documents leaked by NSA whistleblower Edward Snowden indicate the CIA requested $685.4 million for computer network operations in 2013, compared to $1 billion by the U.S. National Security Agency (NSA).
TURNABOUT IS FAIR PLAY?
NBC also reported that the CIA’s Center for Cyber Intelligence was tasked by the Obama administration last year to devise cyber attack strategies in response to Russia’s alleged involvement in the siphoning of emails from Democratic National Committee servers as well as from Hillary Clinton‘s campaign chief John Podesta. Those emails were ultimately published online by Wikileaks last summer.
NBC reported that the “wide-ranging ‘clandestine’ cyber operation designed to harass and ’embarrass’ the Kremlin leadership was being lead by the CIA’s Center for Cyber Intelligence.” Could this attack have been the Kremlin’s response to an action or actions by the CIA’s cyber center? Perhaps time (or future leaks) will tell.
Speaking of the NSA, the Wikileaks dump comes hot on the heels of a similar disclosure by The Shadow Brokers, a hacking group that said it stole malicious software from the Equation Group, a highly-skilled and advanced threat actor that has been closely tied to the NSA.
What’s interesting is this Wikileaks cache includes a longish discussion thread among CIA employees who openly discuss where the NSA erred in allowing experts to tie the NSA’s coders to malware produced by the Equation Group. As someone who spends a great deal of time unmasking cybercriminals who invariably leak their identity and/or location through poor operational security, I was utterly fascinated to read this exchange.
BUG BOUNTIES VS BUG STOCKPILES
Many are using this latest deluge from WikiLeaks to reopen the debate over whether there is enough oversight of the CIA’s hacking activities. The New York Times called yesterday’s WikiLeaks disclosure “the latest coup for the antisecrecy organization and a serious blow to the CIA, which uses its hacking abilities to carry out espionage against foreign targets.”
The WikiLeaks scandal also revisits the question of whether the U.S. government should instead of hoarding and stockpiling vulnerabilities be more open and transparent about its findings — or at least work privately with software vendors to get the bugs fixed for the greater good. After all, these advocates argue, the United States is perhaps the most technologically-dependent country on Earth: Surely we have the most to lose when (not if) these exploits get leaked? Wouldn’t it be better and cheaper if everyone who produced software sought to crowdsource the hardening of their products?
On that front, my email inbox was positively peppered Tuesday with emails from organizations that run “bug bounty” programs on behalf of corporations. These programs seek to discourage “full disclosure” approach — e.g., a researcher releasing exploit code for a previously unknown bug and giving the affected vendor exactly zero days to fix the problem before the public finds out how to exploit it (hence the term “zero-day” exploit). Rather, the bug bounties encourage security researchers to work closely and discreetly with software vendors to fix security vulnerabilities — sometimes in exchange for monetary reward and sometimes just for public recognition.
Casey Ellis, chief executive officer and founder of bug bounty program Bugcrowd, suggested the CIA WikiLeaks disclosure will help criminal groups and other adversaries, while leaving security teams scrambling.
“In this mix there are the targeted vendors who, before today, were likely unaware of the specific vulnerabilities these exploits were targeting,” Ellis said. “Right now, the security teams are pulling apart the Wikileaks dump, performing technical analysis, assessing and prioritizing the risk to their products and the people who use them, and instructing the engineering teams towards creating patches. The net outcome over the long-term is actually a good thing for Internet security — the vulnerabilities that were exploited by these tools will be patched, and the risk to consumers reduced as a result — but for now we are entering yet another Shadow Brokers, Stuxnet, Flame, Duqu, etc., a period of actively exploitable 0-day bouncing around in the wild.”
Ellis said that — in an ironic way, one could say that Wikileaks, the CIA, and the original exploit authors “have combined to provide the same knowledge as the ‘good old days’ of full disclosure — but with far less control and a great many more side-effects than if the vendors were to take the initiative themselves.”
“This, in part, is why the full disclosure approach evolved into the coordinated disclosure and bug bounty models becoming commonplace today,” Ellis said in a written statement. “Stories like that of Wikileaks today are less and less surprising and to some extent are starting to be normalized. It’s only when the pain of doing nothing exceeds the pain of change that the majority of organizations will shift to an proactive vulnerability discovery strategy and the vulnerabilities exploited by these toolkits — and the risk those vulnerabilities create for the Internet — will become less and less common.”
Many observers — including a number of cybersecurity professional friends of mine — have become somewhat inured to these disclosures, and argue that this is exactly the sort of thing you might expect an agency like the CIA to be doing day in and day out. Omer Schneider, CEO at a startup called CyberX, seems to fall into this camp.
“The main issue here is not that the CIA has its own hacking tools or has a cache of zero-day exploits,” Schneider said. “Most nation-states have similar hacking tools, and they’re being used all the time. What’s surprising is that the general public is still shocked by stories like these. Regardless of the motives for publishing this, our concern is that Vault7 makes it even easier for a crop of new cyber-actors get in the game.”
This almost certainly won’t be the last time KrebsOnSecurity cites this week’s big CIA WikiLeaks trove. But for now I’m interested to hear what you, Dear Readers, found most intriguing about it? Sound off in the comments below.