It’s a sign of our current connectedness (and the lack of ability or desire for most of us to live under a digital rock – without an hourly fix of Facebook, Twitter, CNN, blogs, etc – we don’t feel we exist) that the title of this post needs no further explanation.
The Sony “hack” must be analyzed apart from the hyperbole of the media, politics and business ‘experts’ to put the various aspects in some form of objectivity – and more importantly to learn the lessons that come with this experience.
I have watched and read endless accounts and reports on the event, from lay commentators, IT professionals, Hollywood business, foreign policy pundits, etc. – yet have not seen a concise analysis of the deeper meaning of this event relative to our current digital ecosystems.
Michael Lynton (CEO, Sony Pictures) stated on CNN’s Fareed Zakaria show today that “the malware inserted into the Sony network was so advanced and sophisticated that 90% of any companies would have been breached in the same manner as Sony Pictures.” Of course he had to take that position – while his interview was public there was a strong messaging to investors in both Sony and the various productions that it hosts.
As reported by Wired, Slate, InfoWorld and others the hack was almost certainly initiated by the introduction of malware into the Sony network – and not particularly clever code at that. For the rogue code to execute correctly, and to have the permissions to access, transmit and then delete massive amounts of data required the credentials of a senior network administrator – which supposedly were stolen by the hackers. The exact means by which this theft took place have not been revealed publicly. Reports on the amount of data stolen vary, but range from a few to as much as a hundred terabytes. That is a massive amount of data. To move this amount of data requires a very high bandwidth pipe – at least a 1Gbps, if not higher. These sized pipes are very expensive, and normally are managed rather strictly to prioritize bandwidth. Depending on the amount of bandwidth allocated for the theft of data, the ‘dump’ must have lasted days, if not weeks.
All this means that a number of rather standard security protocols were either not in place, or not adhered to at Sony Pictures. The issue here is not Sony – I have no bone to pick with them, and in fact they have been a client of mine numerous times in the past while with different firms, and I continue to have connections with people there. This is obviously a traumatic and challenging time for everyone there. It’s the larger implications that bear analysis.
This event can be viewed through a few different lenses: political, technical, philosophical and commercial.
Political – Initially let’s examine the implications of this type of breach, data theft and data destruction without regard to the instigator. In terms of results the “who did it” is not important. Imagine instead of this event (which caused embarrassment, business disruption and economic loss only) an event in which the Light Rail switching system in Los Angeles was targeted. Multiple and simultaneous train wrecks are a highly likely result, with massive human and infrastructure damage certain. In spite of the changes that were supposed to follow on from the horrific crash some years ago in the Valley there, the installation of “collision avoidance systems” on each locomotive still has not taken place. Good intentions in politics often take decades to see fruition…
One can easily look at other bits of infrastructure (electrical grids, petroleum pipelines, air traffic control systems [look at London last week], telecommunications, internet routing and peering – the list goes on and on – of critical infrastructure that is inadequately protected.
Senator John McCain said today that of all the meetings in his political life, none took longer and accomplished less than cybersecurity subjects. This issue is just not taken seriously. Many major hacks have occurred in the past – this one is getting serious attention from the media due to the target being a media company, and that many high profile Hollywood people have had a lot to say – and that further fuels the news machine.
Now whether North Korea instigated or performed this on its own – both possible and according to the FBI is now fact – the issue of a nation-state attacking other national interests is most serious, and demands a response from the US government. But regardless of the perpetrator – whether an individual criminal, a group, etc. – a much higher priority must be placed on the security of both public and private entities in our connected world.
Technical – The reporting and discussion on the methodology of this breach in particular, and ‘hacks’ in general, has ranged from the patently absurd to relatively accurate. In this case (and some other notable breaches in the last few years, such as Target), the introduction of malware into an otherwise protected (at least to some degree) system allowed access and control from an undesirable external party. While the implanting of the malware may have been a relatively simple part of the overall breach, the design of the entire process, codewriting and testing, steering and control of the malware from the external servers, as well as the data collection and retransmission clearly involved a team of knowledgeable technicians and some considerable resources. This was not a hack done by a teenager with a laptop.
On the other hand, the Sony breach was not all that sophisticated. The data made public so far indicates that the basic malware was Trojan Destover, combined with a commercially available codeset EldoS RawDisk which was used for the wiping (destruction) of the Sony data. Both of these programs (and their similes Shamoon and Jokra) have been detected in other breaches (Saudi Aramco, Aug 2012; South Korea, Mar 2013). See this link for further details. Each major breach of this sort tends to have individual code characteristics, along with required access credentials with the final malware deliverable package often compiled shortly before the attack. The evidence disclosed in the Sony breach indicates that stolen senior network admin credentials were part of the package, which allowed the full and unfettered access to the network.
It is highly likely that the network was repeatedly probed some time in advance of the actual breach, both as a test of the stolen credentials (to see how wide the access was, and to inspect for any tripwires that may have been set if the credentials had become suspect).
The real lessons to take away from the Sony event have much more to do with the structure of the Sony network, their security model, security standards and practices, and data movement monitoring. To be clear, this is not picking out Sony as a particularly bad example: unfortunately this firm’s security practices are rather the norm today: very, very few commercial networks are adequately protected or designed – even financial companies who one would assume have better than average security.
Without having to look at internal details, one only has to observe the reported breaches of large retail firms, banks and trading entities, government agencies, credit card clearing houses… the list goes on and on. Add to this that not all breaches are reported, and even less are publicly disclosed – the estimates range from 20-30% of network security breaches are reported. The reasons vary from loss of shareholder or customer trust, appearance of competitive weakness, not knowing what actually deserves reporting and how to classify the attempt or breach, etc. etc. In many cases data on “cyberattacks” is reported anonymously or is gathered statistically by firms that handle security monitoring on an outsource basis. At least these aggregate numbers give a scope to the problem – and it is huge. For example, IBM’s report shows for one year (April 2012 – April 2013) there were 73,400 attacks on a single large organization during this time period. This resulted in about 100 actual ‘security incidents’ during the year for that one company. A PWC report shows that an estimated 42 million data security incidents will have occurred during 2014 worldwide.
If this amount of physical robberies were occurring to firms the response, and general awareness, would be far higher. There is something insidious about digital crime that doesn’t attract the level of notice that physical events do. The economic loss worldwide is estimated in the hundreds of billions of dollars – with most of these proceeds ending up in organized crime, rogue nation-states and terrorist groups. Given the relative sophistication of ISIS in terms of social media, video production and other high-tech endeavours, it is highly likely that a portion of their funding comes from cybercrime.
The scope of the Sony attack, with the commensurate data loss, is part of what has made this so newsworthy. This is also the aspect of this breach that could have mitigated rather easily – and underscores the design / security practices faults that plague so many firms today. The following points list some of the weaknesses that contributed to the scale of this breach:
- A single static set of credentials allowed nearly unlimited access to the entire network.
- A lack of effective audit controls that would have brought attention to potential use of these credentials by unauthorized users.
- A lack of multiple-factor authentication that would have made hard-coding of the credentials into the malware ineffective.
- Insufficient data move monitoring: the level of data that was transmitted out of the Sony network was massive, and had to impact normal working bandwidth. It appears that large amounts of data are allowed to move unmanaged in and out of the network – again an effective data move audit / management process would have triggered an alert.
- Massive data deletion should have required at least two distinct sets of credentials to initiate.
- A lack of internal firewalls or ‘firestops’ that could have limited the scope of access, damage, theft and destruction.
- A lack of understanding at the highest management levels of the vulnerability of the firm to this type of breach, with commensurate board expertise and oversight. In short, a lack of governance in this critical area. This is perhaps one of the most important, and least recognized, aspects of genuine corporate security.
Philosophical – With the huge paradigm shift that the digital universe has brought to the human race we must collectively asses and understand the impacts of security, privacy and ownership of that ephemeral yet tangible entity called ‘data’. With an enormous transformation under way where millions of people (the so-called ‘knowledge workers’) produce, consume, trade and enjoy nothing but data. There is not an industry that is untouched by this new methodology: even very ‘mechanistic’ enterprises such as farming, steelmills, shipping and train transportation are deeply intertwined with IT now. Sectors such as telecoms, entertainment, finance, design, publishing, photography and so on are virtually impossible to implement without complete dependence on digital infrastructures. Medicine, aeronautics, energy generation and prospecting – the lists go on and on.
The overall concept of security has two major components: Data Integrity (ensuring that the data is not corrupted by either internal or external factors, and that the data can be trusted; and Data Security (ensuring that only authorized users have access to view, transmit, delete or perform other operations on the data). Each are critical – Integrity can likened to disease in the human body: pathogens that break the integrity of certain cells will disrupt and eventually cause injury or death; Security is similar to the protection that skin and other peripheral structures provide – a penetration of these boundaries leads to a compromise of the operation of the body, or in extreme cases major injury or death.
An area that is a particular challenge is the ‘connectedness’ of modern data networks. The new challenge of privacy in the digital ecosystem has prompted (and will continue to) many conversations, from legal to moral/ethical to practical. The “Facebook” paradigm [everything is shared with everybody unless you take efforts to limit such sharing] is really something we haven’t experienced since small towns in past generations where everybody knew everyone’s business…
Just as we have always had a criminal element in societies – those that will take, destroy, manipulate and otherwise seek self-aggrandizement at the expense of others – we now have the same manifestations in the digital ecosystem. Only digi-crime is vastly more efficient, less detectable, often more lucrative, and very difficult to police. The legal system is woefully outdated and outclassed by modern digital pirates – there is almost no international cooperation, very poor understanding by most police departments or judges, etc. etc. The sad truth is that 99% of cyber-criminals will get away with their crimes for as long as they want to. A number of very basic things must change in our collective societies in order to achieve the level of crime reduction that we see in modern cultures in the physical realm.
A particular challenge is mostly educational/ethical: that everything on the internet is “free” and is there for the taking without regard to the intellectual property owner’s claim. Attempting to police this after the fact is doomed to failure (at least 80% of the time) – not until users are educated to the disruption and effects of their theft of intellectual property. This attitude has almost destroyed the music industry world-wide, and the losses to the film and television industry amount to billions of dollars annually.
Commercial – The economic losses due to data breaches, theft, destruction, etc are massive, and the perception of the level of this loss is staggeringly low – even among commercial stakeholders whom are directly affected. Firms that spend massive amounts of time, money and design effort to physically protect their enterprises apply the flimsiest of real effective data security efforts. Some of this is due to lack of knowledge, some to lack of understanding of the core principals that comprise a real and effective set of procedures for data protection, and a certain amount of laziness: strong security always takes some effort and time during each session with the data.
It is unfortunate, but the level of pain, publicity and potential legal liability of major breaches such as Sony are seemingly necessary to raise the attention that everyone is vulnerable. It is imperative that all commercial entities, from a vegetable seller at a farmer’s market that uses SnapScan all the way to global enterprises such as BP Oil, J.P. Morgan, or General Motors take cyber crime as a continual, ongoing, and very real challenge – and deal with it at the board level with same importance given to other critical areas of governance: finance, trade secrets, commercial strategy, etc.
Many firms will say, “But we already spend a ridiculous amount on IT, including security!” I am sure that Sony is saying this even today… but it’s not always the amount of the spend, it’s how it’s done. A great deal of cash can be wasted on pretty blinking lights and cool software that in the end is just not effective. Most of the changes required today are in methodology, practice, and actually adhering to already adopted ‘best practices’. I personally have yet to see any business, large or small, that follows the stated security practices set up in that particular firm to the letter.
– Ed Elliott
Past articles on privacy and security may be found at these links:
CONTENT PROTECTION – Methods and Practices for protecting audiovisual content
Anonymity, Privacy and Security in the Connected World
Privacy, Security and the Virtual World…
Who owns the rain? A discussion on accountability of what’s in the cloud…
Privacy: a delusion? a right? an ideal?
Privacy in our connected world… (almost an oxymoron)
NSA (No Secrets Anymore), yet another Snowden treatise, practical realities…
It’s still Snowing… (the thread on Snowden, NSA and lack of privacy continues…)