Data Theft Notifications - How Soon is Too Soon?

bsdbigot asks: "I started getting a bunch of stock-tout spam in the last month or so. The other day, I happened to look and see it was coming in to an email address I had dedicated to my online trading account account. I've spoken to the online trading company, and I've given them the info on these spams. It turns out there is an 'ongoing investigation,' which includes 'outside agencies,' but they stop short of saying that there is any theft or breach. How soon should such a company let its customers know that their data has been compromised? Should they wait until they have all the details and have plugged the breach, or should they let customers know that there is a possible problem as soon as they recognize it?"

"Personally, I believe a security breach has occurred. So, I asked them how many people are affected by this; they feel certain that it's an isolated problem, because they haven't received a deluge of complaints. They don't know how these spammers got my reserved email address from my online broker (but they didn't sell it, they are quite clear on that), so how can they be so certain it's not their entire database, and how can they be so sure that things like my SSN and bank routing information wasn't also stolen?"

Do more

[omeomi (675045)]

They should do more to keep it from happening in the first place. Seriously, there's a new breach at some major corporation or government office every other week or so. It's ridiculous.

Re:

[bky1701 (979071)]

Damn software pirates stealing data!!!!1!111!11one

Re:Do more

[AusIV (950840)]

That's easy to say, but it's really not so simple. Some data leaks happen because of software issues. More likely an employee figured they could make a buck selling data. Hiring more trustworthy employees requires paying more money, and that has to get passed on to the customers, who in turn take their business somewhere cheaper and less trustworthy. Customers want security, but they're not willing to pay a little extra for it.

Re:Do more

[omeomi (675045)]

Hiring more trustworthy employees requires paying more money, and that has to get passed on to the customers, who in turn take their business somewhere cheaper and less trustworthy

For companies and agencies that have to have highly sensitive information like SSN's on file, there should be an exceptionally small number of people who have access to that information. A small enough number that I can count them on one hand. And none of those people should ever be allowed to take any portion of that list out of the system in any way, not on a thumb drive, not on a laptop, nothing. The vast majority of the employees should only be able to access the last 4 numbers of any given person for varification purposes.

Re:Do more

[houghi (78078)]

For companies and agencies that have to have highly sensitive information like SSN's on file

First start with the fact wether or not the company needs the SSN or not. When in doubt, the answer is no.

employees should only be able to access the last 4 numbers of any given person for varification purposes.

It is a Social Security Number, not a Person Verification Number. If you use it for anything else then for Social Security reasons, you do not need to get it in the first place.
The best way not to loose the data or be tempted to sell it is not to have it.

Re:

[lrichardson (220639)]

First start with the fact wether or not the company needs the SSN or not. When in doubt, the answer is no.

In this day and age, the answer is Yes. Names change - people get married, divorced, decide to use Chuck or Charly instead of Charles. People move. Matching things up on these two - name and address - works 99.9% of the time (with a little effort) - but isn't absolute. SSN (and SIN for those in the Maple Leaf state) allows a match for that final 0.1% percent. (Yes, SSN change occasionally too, but

Re:

[The Snowman (116231)]

Try it next time you sign up for a bank account.

Your bank reports capital gains on your accounts to the IRS. They need your SSN. If you don't give it to them, they probably won't give you an account.

Re:

[Asic Eng (193332)]

Customers are not willing to pay more on the mere chance that the data is more secure. This is a great example of a market not delivering what customers want, because finding out whether one company is better at providing the service than the other, is close to impossible. (I'm joking of course, the market is the only perfect thing in the known universe.)

Re:

[aussie_a (778472)]

That's funny. I know plenty of people who buy doors with better locks then cheaper doors. Guess they're just fictional people though?

Re:

[The Snowman (116231)]

Very true, yet you get a drug test but how many firms do a background or credit check on everyone who comes in contact with the data? Are your contractors liable?

Employees and contractors coming in contact with money, financial data (of which SSN is one piece), and any other customer data should be bonded. That is not a perfect solution, but a good first step. Try working in a bank branch without being bonded -- probably not going to happen. Banks know there's a lot at risk (and the government probably req

Re:

[sumdumass (711423)]

I had a two simular email address that I used only for two seperate things. Everything was fine and dandy untill they started being flooded with spam.

Turns out someone misconfigured a mailserver and someone was able to havest the email acounts for all the domains configured on the server. The other problem was traced to some bullshit E-Greeting card that my girlfriend devided to open when she checked her mail. Several weeks after the problem started on that address, the same E-card she sent me was implicate

Safe/sorry

[sporkme (983186)]

Lock it down. [ftc.gov] Cancel the email account and have any attached credit cards cancelled/changed. Change your checking account number. Keep thorough records and dig to find recent bank statements, etc. This can be a huge hassle.

File complaints with the federal and your state Attorney Generals against the trading company immediately. Consider a 6-month paid monitoring service from a major credit reporting bureau. Both the feds and your state will have advisory hotlines. IANAL and slashdot is not the place you want to go for this kind of information. Basically, don't fsck around if you think anything has been compromised.

I've been there, and these steps cost me a few dollars but saved me tens of thousands. Overseas types are pretty damned creative with your numbers. paranoid != not out to get you.

Re:

[sporkme (983186)]

No, genius, US Attorney Generals do not exactly have jurisdiction over the overseas types --hence the emphasis. Please, don't anyone train US resident scammers! What are you arguing here? Just felt like posting? Did I somehow indicate that I am prejudiced against US Citizen scammers vs. foreign ones? Are you standing up for some group, whose rights I have trampled? I mean, really... bouncing off the walls.

Plug the hole first

[ShaunC (203807)]

Should they wait until they have all the details and have plugged the breach, or should they let customers know that there is a possible problem as soon as they recognize it?

If there's actually a security situation, I'd rather they plug the hole first prior to making an announcement.

As soon as it becomes public knowledge that they've got a vulnerability somewhere, the number of people poking around their interface attempting to stumble upon that hole (or other ones) will skyrocket. Better to fix known problems before they essentially invite the community to look for chinks in their armor. That said, as soon as any known holes are patched, they should inform the affected users; or, if they can't determine whose information was nabbed, they should alert all of their customers.

Keep in mind that no matter how suspicious the circumstances, unless you use that email address solely for your brokerage account, there's really no way to prove a connection unless the company admits it. A friend of mine started playing online poker, used his email address to sign up for the site, and doesn't get any poker spam. A week or so later, his wife started getting a ton of poker-related spam at her email address. It's just a coincidence, though it's about impossible to convince her of that.

I've seen a huge uptick in stock spam lately, across the board (I have a number of email accounts and only one of them is tied to a brokerage). Maybe you're just on the same spam lists :)

D'oh!

[ShaunC (203807)]

I missed the part about "dedicated to my online trading account account." It sounds like there's definitely been a breach, but my opinion of when customers should be notified remains the same.

How stupid is E*Trade?

[Jah-Wren Ryel (80510)]

Here's my story, it meanders off-topic but I think it is worth posting as an example of another kind of data breach, one caused by corporate greed:

Like the article-poster I'm one of those guys who uses individualized addresses for each online entity they deal with, as in slashdot thinks my email is [e-mail], amazon thinks it is [e-mail] and etrade thinks it is [e-mail] - those examples are simplified for illustrative purposes.

A while back, before the bubble burst, I dabbled in some options trading in my etrade account. Therefore, Etrade's marketing department decided that would make my contact information something they could sell to the CBOE and I started getting bi-weekly spam from somebody on behalf of the CBOE trying to sell me all kinds of bullshit options information -- all sent to my etrade-only address.

After about a year of that crap, it finally stopped on its own. But then I started to get spam from the same mailing-list operator that the CBOE had used, but this time they were promoting other brokerages like TD Waterhouse, and most recently "TradeKing" which seems very questionable.

Whenever I get one these brokerage spams, I have to laugh. Etrade breached my privacy to make a buck or two and I'm sure they did the same thing to tens of thousands of other customers. But the end result is that their competition now has a confirmed mailing list of etrade customers, and the stupid greedy bastards GAVE it to them.

I've since opened an account with TD Waterhouse (aka Ameritrade) and make most of my trades through them, in part because of etrade's callous treatment of my privacy. I wonder how many others have done the same...

Re:How stupid is E*Trade?

[(H)elix1 (231155)]

And for those who can't run their own email servers, a handy trick for those using a gmail account is to add a '+' to the user name, and it will deliver. Say I had a gmail account called [e-mail]. I could email [e-mail] and it will resolve to the [e-mail] address. Very handy for finding out who is being bad with privacy information when they ask for an email address.

Re:

[chgros (690878)]

a handy trick for those using a gmail account is to add a '+' to the user name
The problem is that at least 1/2 of the services on the web will consider this an invalid address (despite it being perfectly valid). Very annoying.

Re:

[jfengel (409917)]

I would expect that a spammer would automatically strip out anything after the +, but I don't have any experimental data on that.

Re:

[Asic Eng (193332)]

Another good option would be to use http://sneakemail.com/ [sneakemail.com] You can generate addresses there each time you need to specify an email address - all the mail to these addresses will then be forwarded to your main account. If one of the addresses starts to bring in spam you can just remove it - and you know who caused the problem, too.

Re:

[ccarr.com (262540)]

And for those who DO run your own email servers (and DNS servers) I offer this method: [e-mail], [e-mail], etc. In other words, but the unique part in the third-level domain instead of the user name.

Down sides: creating a new address is more involved; if you don't control your own DNS servers, you have to wait for the zone to reload. I've scripted most of this so that I can set up a new one in under a minute.

Up side: when an address is abused, you just yank th

As stupid as Ameritrade

[DuctTape (101304)]

I've since opened an account with TD Waterhouse (aka Ameritrade)...

Ameritrade/TD-W also let its email addresses out, too. My specifically-for-Ameritrade email address got vanilla (same type as my other accounts; not investing at all) spam. So I changed it. Again.

DT

Re:

[ptbarnett (159784)]

Like the article-poster I'm one of those guys who uses individualized addresses for each online entity they deal with [....]

I do the same thing. So, I'll get to the point quickly...

The email address that I use for my Hertz rental membership has been distributed to spammers, twice. The first time, I sent a complaint and after a while I got a patronizing response about how it couldn't be them, and was instead someone else to whom I had given the address. It must have been a form response, as I had alr

Maybe YOU were hacked

[mikem170 (698970)]

Objectively you did not prove how the spammer got your email address. It might have been a breach at the trading account company. It might have been someone capturing traffic at your ISP.

Re:

[Asic Eng (193332)]

The trading company might also have given out the address voluntarily (and now doesn't want to admit to that) or it could be a lucky guess of the spammer (maybe a dictionary attack of sorts). I know they used to try use commonly-used nicks on my domain for a while. (Then I turned the catch-all off...)

Re:

[ptbarnett (159784)]

Right! Someone has been capturing traffic to his ISP, and instead of grabbing his credit card info and his passwords, they are just content to snatch his e-mail address. If he was hacked, the hackers would have done lot worse than stealing his email addy you dumbass!

Credit card info and passwords will be encrypted, if a user is taking any precautions at all.

However, most connections between MTAs (Message Transfer Agents like Sendmail, Postfix, etc.) are not encrypted. My Postfix server offers TLS to a

This has happened to me before...

[heldlikesound (132717)]

I create addresses specifically to receive mail from retailers I order from. For example: [e-mail].

I NEVER type these addresses anywhere, and they are not something a wide net spam sender would guess...

Over the last few years i have had about 4 situations where those very account specific addresses began receiving a LOT of spam.

The sites included Dell, and PCMall. The PCMall ones very primarily sexual in nature...

I have thought of every possible way they could have gotten that addre

Re:

[Propaganda13 (312548)]

I used to do the same thing with Sneakmail(ranomly generated disposable email addresses). These spam floods aren't necessarily from breaches, but from sharing with partner companies. Companies' information policies are subject to change, and once your info starts flowing to other companies, it's hard to control.

Re:

[EtherMonkey (705611)]

I NEVER type these addresses anywhere, and they are not something a wide net spam sender would guess...

You think that nobody has ever come up with this idea before, creating unique mailboxes for various relationships? I've been doing it for 6 or 7 years, and I've taught dozens others the same idea. I probably got the idea from someone else or an article I read online or in print. But regardless, it is a simple and logical scheme that some savvy spammers are bound to figure out on their own (more so now

Dictionary attacks?

[spinfire (148920)]

As an operator of a mailserver I know I do frequently get dictionary attacks (searching through names for mailing addresses) and sometimes these turn up addresses which aren't used (like my "stats" account or mail sent to "apache" or "mailman"). Usually these addresses later received subsequent spam - and often it is the most shady kind of spam such as the stock scam emails. So it is possible that the address may have been discovered through this means.

Are we talking about Ameritrade?

[SysKoll (48967)]

"I started getting a bunch of stock-tout spam in the last month or so. The other day, I happened to look and see it was coming in to an email address I had dedicated to my online trading account account. I've spoken to the online trading company, and I've given them the info on these spams. It turns out there is an 'ongoing investigation,'

Is the trading company called Ameritrade by any chance? They got a leak problem, maybe an insider job. Look at this thread on spamgourmet (an anti-spam site that I help with): http://bbs.spamgourmet.com/viewtopic.php?t=81&star t=60 [spamgourmet.com]

Re:

[dangitman (862676)]

[spamgourmet.com]

Damn, I went there looking for recipes. Please stop using misleading domain names.

Immediate but reserved

[mikesd81 (518581)]

Victims should be told right away that it is suspected that there was a breach, and outlined how an investigation will be performed. But it should be careful (mostly for legal reasons) that it be reserved and only enough information be given.

certain laws may apply

[sharp-bang (311928)]

In the banking industry, the applicable regulation [federalreserve.gov] is fairly strict... the institution must "promptly" notify customers of a material breach and there are relatively few loopholes. So if your broker or whoever was part of a bank, then this would apply. However, if your e-mail address was all that was compromised, they don't really need to notify you. By definition, e-mail addresses are not private information, any more than your physical address is. A number of states, notably California [ca.gov], have privacy laws that can be invoked, but the trigger for a material breach is usually the compromise of a combination of personal identifying data such as name and address (including e-mail addresses) and sensitive nonpublic personal information such as login credentials, account numbers, etc. You might see whether there is a law in your state [consumersunion.org] that applies.

Oops

[sharp-bang (311928)]

assuming that you live in or are doing business in the USA

How soon is too soon???

[st0rmshad0w (412661)]

Before it happens?

Sometimes hard to tell

[dtfinch (661405)]

If a tape is missing, how do they know if it's been stolen? If a system is infected with a general purpose trojan, how do they know the extent to which the data was compromised, or if anything was downloaded at all? There would be a lot of false alarms if companies had to alert customers every time there was a possibility of data theft.

But if you believe that sensitive data was probably stolen, then you should have to alert the people you believe were probably affected immediately. The only problem with thi

Re:

[dtfinch (661405)]

I suppose that when data theft is confirmed, a fine could be levied in proportion to the time it took between the actual theft and its publication. That would encourage companies to report probable thefts, if the threat of fine is severe enough, while allowing them to keep the less probable breaches secret.

It's a balancing game...

[ChePibe (882378)]

Sometimes, a company can only make a security threat worse by declaring the problem exists.

Let's take a stolen laptop, for example. If Company A's suffers a laptop theft, and the laptop (for whatever stupid reason) has the personal data of thousands of customers or employees on it, how should that company respond? This is obviously an example of poor security to begin with (no one should have that kind of information on a laptop taken off the premises), but how do you keep a bad situation from getting wor

Re:

[laughingcoyote (762272)]

I'm not sure there's always a best way to handle these things - sometimes it could be informing everyone, at other times it could just mean scrutinizing accounts more closely while keeping everything quiet. It's a hard thing to balance.

The same rule as always applies here-SECURITY THROUGH OBSCURITY DOES NOT WORK. You could be dealing with a couple of punk kids who randomly stole a laptop and are off at the first opportunity to pawn it-or you could be dealing with organized gangsters who know damn well w

Are they incorporated in California???

[soren42 (700305)]

So, according to Bill AB 424 [ca.gov] in the Great Sovereign State of California, any company negligent in the protection of customer identity data must immediately inform the offended party upon being made aware of the breach.

I understand that there have been several attempts to leverage that law on behalf of US citizens who can't afford to live in California (us poor, ol' east coast folks!) to require major corporations transacting any business in California to immediately disclose based on that law.

I'm sure there's jurisdictional issues, but there's at least some chance in hell that virtue jurisprudence will prevail.

Anyone with an actual Litt.D, SJD, or otherwise more qualified care to add fact to my hype and speculation? :)

Re:

[sharp-bang (311928)]

32 states have similar laws [consumersunion.org]. Disclosure of identity data *only* may not be sufficient cause. But if you think there's an issue and you're in the proper jurisdiction, a letter to the firm copying the state attorney general might be helpful.

Transfer your account to another broker. Now.

[Animats (122034)]

Get out of that broker now. Move all your assets to another broker. You don't want to have assets with a broker in trouble.

I've been through a broker bankruptcy, and it's a huge hassle. Yes, you eventually get the assets back, but you may be trapped in a position and unable to trade out of it.

priorities

[macadamia_harold (947445)]

How soon should such a company let its customers know that their data has been compromised?

that depends, how long does it take to finance a new ferrari and a yacht to ship it out of the country?

Notify Immediately

[ErichTheWebGuy (745925)]

I bought a CD from an online store a few years back. They got hacked, and customers' credit card numbers were stolen. I got a call that same day from the store, saying that they were aware of a problem and that I should take measures to protect myself. I really appreciated that. I have gone back to them several times, because of their honesty with me, and also because of the borderline-paranoia about security that follows a successful attack/theft.

Re:

[DerekLyons (302214)]

I bought a CD from an online store a few years back. They got hacked, and customers' credit card numbers were stolen. I got a call that same day from the store, saying that they were aware of a problem and that I should take measures to protect myself. I really appreciated that. I have gone back to them several times, because of their honesty with me, and also because of the borderline-paranoia about security that follows a successful attack/theft.

Myself I'd run screaming from such an online store, and war

ANSI and BBB Standards

[joeflies (529536)]

Although this was JUST announced a few weeks ago, ANSI and the Better Business Bureau [ansi.org] are setting up a working group to define standards and best practices for how to address identity theft. The scope is to first catalogue what standards and best practices exist, and then go beyond and define what else needs to be documented.

Whether or not this results in the answer to your question (how long notification should be given), at least this is a step in the right direction for some centralized thinking instead of everyone doing it on their own.

Why should they care?

[Asmor (775910)]

How soon is too soon? At all. For them, at least. There is no real reason for them to admit anything. They don't really lose a whole lot by not admitting things. A couple savvy users isn't really worth the cost of the bad PR. Yeah, ideally they'd let everyone know as soon as the possibility of a leak was made known to them, but this world doesn't run on ideals.

True security only comes when it's in the best interests of the person for whom the security is a cost, particularly at a corporate level. I'm sure t

disposable addresses

[jqh1 (212455)]

There's a long-running thread [spamgourmet.com] on the bbs for spamgourmet discussing a bunch of events like this -- spamgourmet users generally use a unique email address for each of their accounts, and so can quickly identify a problem (unless it was with spamgourmet itself, of course, but records so far show that hasn't happened). The response of the companies varies from complete denial and reticence to surprising accountability. None of it ever ended up in court, afaik.

Re:

[Who235 (959706)]

So, if your email address was "[e-mail]", it is entirely possible that a spammer synthesized the address. You need to use addresses that are not easily guessable, for example, [e-mail]

I agree.

Change your address to your address. . .