The bigger issue (which I think you're getting at) is whether this trade-off accurately accounts for externalities. Even if a security vulnerability only costs Sony $1 million, it could cost their customers, credit card companies, and any number of third parties a lot more than that.
Might be a valid strategy if hackers only had the skills, motivations, and attention span of a grizzly bear.
If the company is a "good citizen", then an approach like that may be justifiable. If you are BP, maybe not so much...
Even if you assume away any problems caused through either ill intent or simple not being perfectly omniscient by lawyers, judges, and jurors (all of whom can cause failures in the civil legal system), the impossibility of that system being a complete mechanism for recompense lies, among other places, in the fact that it is possible for the liable party to have caused harms to the injured party that the liable party is not capable of compensating. The harms a person is capable of doing are not limited by their capacity to pay -- someone with no assets can do an injury causing millions of dollars of property damage.
Lawyers are just the _means_ by which those people exploit the system. They're also the means by which people who are wronged can get justice, so they're exploitation agnostic.
(The problem, of course, is that there are cases where the stakes are non-monetary, or the plaintiff places a much greater value on a relatively small amount of money than a lawyer would...but I don't think that's really what we're talking about here.)
> Spaltro offers a hypothetical example of a company that relies on legacy systems to store and manage credit card transactions for its customers.
The full quote:
> Although Spaltro declines to talk about Sony’s security practices, he says that while Sony Online Entertainment is fully compliant, every company weighs the cost of protecting personal data with the cost of what it would take to notify customers if a breach occurred. Spaltro offers a hypothetical example of a company that relies on legacy systems to store and manage credit card transactions for its customers. The cost to harden the legacy database against a possible intrusion could come to $10 million, he says. The cost to notify customers in case of a breach might be $1 million. With those figures, says Spaltro, “it’s a valid business decision to accept the risk” of a security breach. “I will not invest $10 million to avoid a possible $1 million loss,” he suggests.
http://www.cio.com/article/2439324/risk-management/your-guid...
Not an idiot: Spending $10M on insurance to offset a risk of $1M.
Idiot: Believing that the only cost of total loss of control over personal data is the cost of notifying people you lost control over their data.
With a bit more thinking he could know that 'notifying people' was practically "free" since there are a number of interested parties who will do that (news media, credit card agencies, federal agencies).
But what it really sounds like from that quote is that he has constructed a model in his head of what sort of data they have control over (or not) which is not very representative of what data they actually had, and control over.
So with that said, if the cost is report is 1 million, and the system is vulnerable, and they expect to get compromised then it's not just going to cost them just 1 million. It's going to cost them at least 11 million, or more.
Further, the "don't dix, just accept the risk" costs also won't just be 1 million, it's going to be 1 million every time it's compromised until they fix it. Since he's only accounting for the cost to report the breach, and the box is STILL owned and STILL vulnerable, it's going to keep getting compromised. So its 1 million times each event, which is a potentially infinite since nothing's changed to stop the next event.
To stop that infinite cycle, he has to pay to fix it, or keep paying to report. Either way he's paying more than 1 million. And since Sony I'm sure isn't that stupid, let's assume someone with common sense will order him to stop accepting the risk and the bleeding, then his real risk cost is 1at lesst 1 million to report and fix once that decision is made (and fixing is probably going to cost more now that the original 10 million since the system is now compromised too).
I'm not sure if he realizes this or not, or if it was a clumsy attempt to explain risk, but at least in the statements reported that's a critical flaw in his reasoning. And that's why what he said isn't just unconvincing, it's wrong.
If that's how Sony makes risk management decisions I have a Pinto to sell them.
Who even hires these people?
* know what you're doing with regards to hiring
* don't know what you're doing, and deferring to someone competent to make that hiring decision for you
I know understand why I'm not a CEO.
* an inflated sense of your own value
* competent enough to make other people believe it
Or, as hga proposed:
* be a useful mark to take the fall when it comes
I believe he was promoted because they wanted him away from his previous job, managing a software development department.
Hindsight is a wonderful thing.
An interesting question is this: You're the CISO of a major Hollywood studio. You have a budget of $xx million. Smart, well-resourced people are determined to destroy your company. What do you do?
Both customer data and employee data must be held under these rules.
https://www.gov.uk/data-protection/the-data-protection-act
https://www.gov.uk/personal-data-my-employer-can-keep-about-...
You could construe that he would be instructing his employees to act illegally if he was a UK CIO.
I don't think you could construe that at all. While everyone is making great hay about his statement, the truth is that every organization makes the compromises that he mentions. While perhaps duplicating what andrewfong said, if security is supreme-
-there is zero worker mobility. Want to work at home, or bring the laptop on the flight? Security threat.
-there is zero internet connectivity. Want to browse at your desk? No chance.
-email attachments? Filesharing? Heck no.
-users have absolutely no rights on their workstation beyond the most rudimentary.
-application sets are extremely restrictive and vetted.
-Want to work in the conference room? Call the security center to configure the IPSec, MAC and port filtering security to allow you to move your locked down desktop.
And on and on. While quite a few on here seem to present the notion that they comply with all best practices, extraordinarily few do (even the NSA doesn't). And those best practices cost in both productivity, worker enjoyment, and direct expenditures.
It is never, and has never been, as easy as security versus no security. Against a dedicated, resourced foe, just about anyone and everyone is vulnerable.
A judge may well agree that it is unreasonable but he cannot use the financial loss to the company as the only metric to make that decision.
The HN hive mind victim blaming at it's best.
If you bother to read the proper statement it's specifically about one 'hypothetical' database getting breached.
"The cost to harden the legacy database against a possible intrusion could come to $10 million, he says. The cost to notify customers in case of a breach might be $1 million. With those figures, says Spaltro, “it’s a valid business decision to accept the risk” of a security breach. “I will not invest $10 million to avoid a possible $1 million loss,” he suggests."
And it's a fair statement. You can spend ever spare dollar on IT security and still not be 100% secure, just poor with a crippling system that grinds productivity to a halt. You can see this with the auditor living in gaga land asking employees have passwords with "random letters, numbers and symbols"
It is about risk analysis as he had the courage to say out loud and it is what everyone in real world IT security knows.
Budgets are finite. That's a fact of life. The trick is balancing the risks and hardening the important things so that, on the balance, you have the most security possible for that budget.
Not spending as much money hardening the (hypothetical) legacy DB with a noncritical data of some sort, so that you have more budget to harden the financials database; that is the right decision.
Of course, it doesn't look like that balance was struck effectively very often at Sony Pictures, but the theory is sound :)
http://www.cio.com/article/2439324/risk-management/your-guid...
where he basically advocates cutting corners and doing the bare minimum. It shouldn't be the security executive's job to save a buck and brag about it, yet this is probably the precise mentality that allowed him to move up the ladder.
How'd that work out for you, Sony?
I'm about to walk away from the corporate IT world because I'm over beating my head against the desk while people who literally can not turn a computer on without a 1 hour demonstration tell me they know better.
That was the intent with Massachusetts PII law 201 CMR 17 which stipulates a $5000 fine per incident. I've always heard this interpreted as "per user record exposed" though apparently that's in dispute. $5k per exposed user could add up, but so far I don't think the law has had much impact. (btw it applies to anyone holding MA resident's data wherever the company is located / data is stored). California has a similar (earlier) law I believe. Perhaps we need this to be adopted at the Federal level.
I also like the idea of outsourcing the security/storing of data. Then we'd have companies that become real experts at that, and all the "idiots" would default to using those companies, instead of doing it themselves, poorly.
This isnt their thinking. Industry wide belief is that CIOs role is minimizing financial liability, NOT maximizing security.
something like this: http://www.washington.edu/research/centers/126 http://www.pce.uw.edu/certificates/information-security-risk...
Where professors are >50 year old bumbling fools proudly proclaiming they know f'all about technology, and whole thing revolves around staying compliant (on paper) with federal law (SOX etc) and industry standards (PCI).
I wasted time taking online versions of those courses. Wrote about it previously here https://news.ycombinator.com/item?id=7353205
"CIOs are actively discouraged from possessing any IT background. Just look at the retardation UW is teaching:
https://www.coursera.org/course/inforiskman
https://www.coursera.org/course/inforisk
https://www.coursera.org/course/infosec
Those "people", they call themselves teachers/professionals in the field, will tell you that Information Security is ALL about covering your ass and shifting blame to a third party. All three courses are about getting a slip of paper that says you delegated responsibility clearing your company from any lawful obligations in case of a breach. Not a single peep about technical aspects or real security of data/infrastructure. Its all about getting that "we paid company X to do it for us" waiver.
There was (deleted now) a huge thread on Coursera forum titled Disappointment covering this topic and demonstrating total lack of professionalism or any real world knowledge. Whole UW program is a government funds skimming scheme. Just look at this garbage:
"Center of Information Assurance and Cybersecurity at the University of Washington, designated by the NSA/DHS as a Center for Academic Excellence in Information Assurance Education and Research"
I highly encourage you to check out those courses (I think there is preview available for old lectures) if you are in need of a laugh, or reality check concerning state of CIO education."
Like EU data protection legislation? This has the added benefits of discouraging speculative accumulation of personal data by all and sundry.
Lets say that after Nuremburg, we are not willing to completely absolve low-level players because we realize after the Holocaust that widespread complicity is the only way such an atrocity can happen.
All the same, we aren't assigning the same level of responsibility to the prison guard that we do to Goebbels, etc. We will apply it proportionately.
If you take the Sony crimes and relate them to the Holocaust -- for nihilistic internet argument purposes -- you end up with some small fraction.
Following the Nuremberg methodology, you take that small fraction and apply it 100% to decisionmakers and proportionately to low-level players. The resulting blame is small enough relative to their breach of privacy that the cyber attackers are in the wrong.
Therefore, taking this argument back to Nuremberg does not serve the purpose of justifying the middle managers/analysts breaches of privacy.
Sony on the other had is a fully functioning company that is by far not defeated. Each employee needs to take fully responsibility for the company they work for.
Not sure why people have it in for Sony though.
Examples:
http://boingboing.net/2005/11/14/sony-anticustomer-te.html
http://arstechnica.com/uncategorized/2005/08/5165-2/
http://www.techhive.com/article/240213/sonys_no_sue_playstat...
Jason Spaltro, then executive director of information security at Sony Pictures, called it a "valid business decision to accept the risk of a security breach" in a 2007 interview with CIO Magazine, adding he would not invest "$10 million to avoid a possible $1 million loss."
So basically their thinking was that getting hacked was just the cost of doing business. Of course, they are now discovering that the cost of a really serious hack is much higher than they thought it was.
Which makes me wonder if, at some point, we're going to have to have some kind of controls on who can legally hold personally identifiable information on their systems and who cannot. Right now pretty much anybody can, regardless of whether they're competent enough to protect it. And as a result there's a huge volume of critical information out there stored on systems that are either poorly secured or whose admins have decided, like Sony's, that the ROI on real security is too poor to justify having any, which creates a target-rich environment for hackers to take advantage of.
Attaching serious liability to holding data on systems that aren't secured, or requiring proof of competency/minimum-acceptable-effort in order to avoid such, might shift the ROI calculation on security enough to convince even idiots that it's worthwhile; or, at least, that it's better to outsource holding the data to someone who knows what they're doing (and is willing to back that up by accepting liability) than it is to keep everything in-house on a dusty Windows NT4 box under someone's desk and just cross their fingers.
We already sort of do this sort of thing for financial information, via PCI; but the universe of "data that could do serious damage if it got loose" is much larger than that which PCI covers, as this hack demonstrates. So I wonder how many of these types of giant hacks people will be willing to accept before they start calling for some kind of protection.