Slashdot Mirror
People Suck at Spotting Phishing
JohnGrahamCumming writes "Initial results at SpamOrHam.org show that people don't fare well when trying to spot spams and phishes. This blog entry shows some actual spams and phishes that people fell for, as well as genuine messages that they think are spam." The thing about these s[cp]ams is that they must work sometimes. When I see the messages, I can't fathom 'how'.
At what point in history was this not a problem? Can't say I'm surprised...
- For every action, there is an equal and opposite criticism.
I've seen more sophisticated phishing examples by far, and some are indistinguishable from what might be the real thing. The distinguishing factor from a genuine missive is the best phishes have links to bogus addresses (sometimes denoted with only an IP address), and the destination site asks for information company's won't ask for from an e-mail.
One of the best phishes I've seen was sent to me -- it was ostensibly from my phone company, and it described a problem with my on-line bill pay (I don't). The letter was nicely formatted with the colors and icons of my phone company. The link was a giveaway, when I rolled over it, I could see the IP address, not a phone company web-site.
I researched this a bit more, went to my phone company's web site, and downloaded their graphics. A bit-for-bit comparison of their icons, etc., and the phishers showed them to be identical. (Interestingly, this puts phishers also in the position of being guilty of more crime: copyright violations.)
Had my suspicions not been raised by the fact I wasn't participating in on-line bill pay and the phish indicated that problem, and had I not seen the IP address by rolling over the link (which I only did because of above suspicion), I easily could have been convinced I was dealing with a real e-mail (NOTE: this was two years ago, before phishing had become real big, and it was my first incident.)
I can easily believe many, if not most could fall for well crafted phishing expeditions. I would agree with the cited article, those are weak examples unlikely to catch savvy users (though they still could catch the naive, of which there are millions!). (And, I would claim some of the examples really are nothing more than SPAM.)
...there is no patch for human stupidity.
Most users just don't know better, despite best efforts to educate them otherwise, or make the scams obviously fradulent. Ever seen that 'MSN will never ask you for your password!' type banner on things? Know how many people retain it? Very few.
Informatus Technologicus
He finds it strange that people called that message from "Keith" to be spam... but the thing is, if you have no idea who "Keith" is, it probably IS spam... and if you do know him, you probably would not mark it as such.
The same goes for the US Airways thing. Yeah, it's an example of "not spam", but if you haven't recently bought a US Airways ticket, then the save bet would be that it is.
Oh... and the nun joke is fucking hilarious. That alone made TFA worth reading.
Information wants to be anthropomorphized.
TFA seems to be using a funny definition of spam.
Most would say it's unsolicited commercial junk mail, but he seems to think it means "phony" email. Apparently he doesn't mind receiving weekly airfare specials containing choice bits like "BID FOR TICKETS TO THE BIG GAME IN THE BIG EASY!"
Also re phishing: I'd say paypal is largely at fault for this. They do (did?) send an awful lot of useless mail full of clickable links - they were just begging to get phished because people were so used to receiving authentic but useless clickable mail from them. None of my other banks have done this (although one sends a fair amount of crap not specific to my account - rates and such).
So what if someone thinks a legitimate email from a bank is a phishing scam? Banks shouldn't be using email for anything serious because it makes their customers more susceptible to fraud. If people expect to receive legitimate and sensitive communications from their bank via email, it's that much easier to fall for it.
For example, I got one this morning talking about my home loan account with a large bank I don't have an account with. I know it's a phishing scam just from the From and Subject lines. However, if my own bank sent an email talking about my actual mortgage, I'd treat it in exactly the same way. There's no benefit to giving an email the benefit of the doubt. If there is something my bank needs from me, they can send a letter and I'll go to my local branch to take care of it in person.
Let's say I handed you an entire crate of auto parts, and told you that some of them may be genuine parts, while others might be knockoffs. I give you a whole binder, filled with instructions on how to differentiate between all the different "good" and "bad" parts. Some of these knockoffs are obvious fakes; others are quite cleverly done, requiring you to check for minute details such as whether or not inner surfaces are well-polished, or subtle discrepancies in serial number schemes and product logos.
At what point do you just start winging it? After one day of studious sifting? After a week? A month? When you see a part that you're pretty sure is genuine, but would need to haul out the manual for ten minutes' worth of cross-checking part and serial number ranges to confirm this--at what point do you simply go with your gut?
When somebody who knows what they're doing goes about trying to hoodwink your typical individual, it can be very hard for the individual to know when they're being hoodwinked, even if they know they might be being hoodwinked. It's part of human nature--there's a point at which you just throw your hands in the air and grant your trust to an unknown entity, because it's too tedious or time-consuming to check everything out. Given the average person--heck, even a person who knows a fair amount about the subject--there'll be a point where they just take the damn part and have it installed in their car, because they just want to be done with it and get on with their life. It's the same thing with phishing--unless you're one of those few individuals who has fairly advanced knowledge on the subject, you're eventually going to give up and make a gut-reaction decision to whether or not you "trust" the email you just got, simply because it's more trouble than it's worth to actually dig through it.
Obliteracy: Words with explosions
As someone said, think of how stupid the average person is, and remember half the people are even stupid than that. People suck at spotting when they're being cheated or lied to, which is why phishing, advertisers, and politicians merrily thrive.
For pete's sake people, if you have to show genuine emails, try at leat to sanitize them a little. Some of the 'ham' emails shown still have the full contact information, including the original email address. That's what I call dangerous!
If you don't believe me, go to the web site, and try classifying some emails... You'll see what I mean...
The right to offend is far more important than the right not to be offended. (Rowan Atkinson)
Ah, and by the way, there are many people falling for fraud on the front door. We really shouldn't allow people to open the front door if they have not passed a test or three. The fact that anyone is able to open his front door means that there are a lot of ... uninformed people out there.
The Tao of math: The numbers you can count are not the real numbers.
Mind you, I think that that type of phish is the most sophisticated type of phish, being both elegant and simple. I "fell" for one of those back in the day, in that I got an email from my bank, and it notified me of some account change, so I immediately and without checking the validity of the link on the email...called my bank on the phone and said, "What the hell is up with this?"
They of course, didn't know anything about it, I checked the link and realized it was false. That was just long term ingrained habit that puleld me out of that one, because it was an excellent phish. But how do you teach those habits of suspicion to a layman?
It's just a security issue. I deal with passwords all day every day, and people are awful with their password security. It just doesn't make any sense to them, and they all think that the consequences for this or that little security breach are harmless, and so when something like this comes along, they fall for it, hook, line, and sinker.
ad logicam Claiming a proposition is false because it was presented as the conclusion of a fallacious argument.
I treat all of those emails as a phishing attempt. If I think it has the possibility of being legit, I type in the appropriate web address (no, I don't cut-n-paste, I type in the previous login site), login and verify the contents.
That might be a little harsh. We're seeing increasingly sophisticated phishing stuff -- right down to building a look-alike site of the bank which they are pretending to be.
I think it's getting increasingly difficult for even people who know what they're looking for to spot.
Yes, people need to learn the basics of how to spot and avoid spam and phishing. But, the increasing sophistication of the bad guys makes it a difficult thing to always identify.
Cheers
Lost at C:>. Found at C.
Evil will always triumph, because good is dumb.
Proud member of the American Non Sequitur Society. We might not make much sense, but boy do we love pizza!
Email clients and servers need to start automatically looking at the chain of IP addresses or domains in the headers, and rating them accordingly.
If any header lies, e.g. IP address mismatches with domain name, or two successive Received-by headers don't have consistent information, then RED ALERT.
If the From domain doesn't appear in top-most received line, YELLOW ALERT. If it doesn't appear in any line, RED ALERT.
If the top-most received line's address is from a known spamming domain or open relay, RED ALERT.
If any previous mail-server, such as your ISP's, tagged the message with YELLOW or RED alerts, your alert should be at least this high.
Note that red and yellow alerts don't necessarily indicate spam. They are simply one of many indicators of spam, and should be used as input to the spam/ham decision-making process.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
I don't see how you could possibly think that the results of such a website could be meaningful. Spam filtering is a contextual process. This site cripples the critical component that allows humans to behave differently from naive filters, i.e. judgement based on memory. The claim being made here is that humans can't identify other people's spam (and this makes sense, how can you tell if you're shown a random email whether it's unsolicited or not? the only way you can is by knowing whether the recipient had been signed up for a mailing list or not!). You should NOT conclude, based on that fact, that humans are bad at identifying their own spam.
There are lives at stake here!
Stop using HTML or convert it plain text and it's hard not to spot a phish.
UNIX/Linux Consulting
I conduct almost all of my business online and I don't think this is necessary.
I am never, ever asked for a password or identifying information via email. At least never by the legitimate company.
And I never click a link in an email. If my bank/company wants me to update my information, I type their website URL by hand into Firefox, log into my account section, and do what I need to do.
It basically comes down to this: Don't click links in email.
This one basic rule really does solve 99.999% of all scam problems, while allowing you to conduct business online safely.
Yes there is.
In other news, 50% of people have below-average intelligence.
Jokes about statistics aside, people falling for phishing is our fault. Our fault as in our industry's fault.
We've spent so long training our parents, help-desk clients, and other tech-stupid creatures that the way to respond to mysterious dialog boxes is to "Just click OK!" that at this stage the damage is essentially permanent.
Their natural instinct was to treat computers with suspicion, and we beat it out of them.
Yay for us.
I was reading a Dilbert strip there recently where the PHB was interviewing candidates by showing them his junk mail and asking them what they would do with it.
Another couple of candidates and he would get through his inbox.
There's an intense feeling of Deja-Vu here.
Genesis 1:32 And God typed
Anyone spotted red text "TRIAL COPY" across the titlebars in the screenshots?
Looks like a "feature" of some screenshot capture shareware.
Nevertheless, I think (having in mind the topic of TFA) this doesn't add them much credibility.
Rediculous is ridiculous!
The weight of getting the word out about these things to the average user is going to need to lay on someone, probably ISPs. It should be one of their responsibilities to attempt to keep their users safe. We can educate people about some of the basics, watching out for links that are just IPs and etc (and thunderbird already has some features regarding this), but some of the higher level checks need to be done automatically by software.
But no matter how sophisticated filtering technology gets, the numer of ways that data can be manipulated and the sheer volume of traffic means that some of these things are going to get through. And while extra IQ points don't automatically confer amazing powers, they might allow people to become more suspicious of something that doesn't "look quite right."
We've been educating people from the mid-60s on that smoking is deleterious to their health. Has smoking ceased? No. In that case, it's the addicition to nicotine and the idea that smoking someone cool (ever kiss a smoker? Yuck!). In the case of email, I'd be willing to bet the vast majority of folks who click on these links in phishing emails are: 1) paranoid types, who have so bought into the identity theft idea they can't stand the thought that someone may be stealing their identity, 2) people who have little sophistication in general and virtually none in the world of PCs and the Internet, 3) greedy people, who thin that they'll just fire off a few hundred dollars of their hard-earned money and some friendly guy in Nigeria will make them rich, or 4) lonely folks who just want to talk to someone or feel a part of something.
GetOuttaMySpace - The Anti-Social Network
I completed about four tests before I started to get the feeling that I was actually working on training their filter. I felt like I should be charging a fee. Most of the tests are bogus. One email asked me to add some addresses to the "TW mailing list". I don't have context - in this scenario, do I work for an employer who has a "TW mailing list"? Do I manage it? The answer has everything to do with the way I'd rank it. In fact, most of the emails referred to specific people, and knowing or not knowing them would control the rating on the email.
EVERY serious site has a disclaimer stating they will NOT ask you for your details by email. EVERY scam involves them sending an email asking for your details.
In the early days, yes. Now, many phishers have wised up. They'll send you a phish that, save for one or two links, looks absolutely legitimate. You click the link, it sends you to a page at ebay.verification-department.com that mimics an actual eBay login page. You'll "log in", then they'll welcome you and very professionally gather your information--all, of course, after you've "logged in" to their system.
You can't cheat a honest man
Oh, you most certainly can. Just 'cause something rolls off the tongue nicely doesn't mean it's true.
and you can't phis a person who thinks.
Again, we're talking about attrition and trust. Unless you have a quite solid understanding of what phishing is, how to identify it, and how to go about avoiding it, you're going to eventually just trust something that looks legitimate enough. It's simply not feasible to expect that every single user of email will have enough technical know-how to identify and avoid getting phished.
You've got telephone slamming, you've got phishing, you've got insurance fraud, you've got pyramid schemes, you've got con artists--if we were all simply smart enough to know a rat when we saw one, none of these would be a problem. The problem is that many, many people have ductile minds and want to trust other people. If you're somebody who is willing to cheat another person out of their money, odds are that you'll eventually nail somebody. It's attrition, plain and simple--eventually, people simply let their guard down, even if only for a moment.
Obliteracy: Words with explosions
Rule 1: It's almost certainly not legit, before you even look.
Rule 2: If it seems legit, then go to your browser and manually go to the institution's website and log in normally, do not use hotlinks provided in any email.
My rule 1 used to be just "it's not legit" - none of my financial institutions EVER contacted me via email up until about 6 months ago. Now they do, so I've modified it a bit.
You'd think people would get a BIT of a clue from the fact that, like me, they must be getting very valid-looking emails from places that they don't even have accounts with. You'd think that would tell them something.
unfortunately, there are problems with that as well - there are some legit sites that will redirect you off of their main domain, sometimes even to an IP address. Insane? Yes. But it happens. So for people who actually DO know what the hell they're doing, the problem isn't phishes that look like real sites, it's real sites that look like phishes.
John Graham-Cumming says that the Travelocity email at the bottom of the his blog essay "really is a genuine message from Travelocity and not a spam."
I beg to differ. I have no problem believing that it "really is a genuine message from Travelocity."
But spam doesn't mean "phony," it means "unsolicited commercial email." (And in my own opinion that includes "unknowingly 'solicited' commercial email.")
In order for Graham-Cumming or anyone else to say that Travelocity email is not spam, they would need to know whether it was solicited. You can't tell by any examination of the message itself.
If it was actively solicited by someone specifically checking a box requesting to be notified of offers, then, sure, it's not spam. If it was opt-out spam with the opt-out option hidden... or implicit... then it darn well is spam.
Mostly likely this particular email is in a grey area... quite likely an opt-out was plainly visible, but needed to be actively chosen, at some point in the travel booking process where a customers thoughts are likely to be elsewhere (where IS that security code on the back of my credit card?).
But it is absolutely wrong to stay that the Travelocity message is "not spam," just because it is really from Travelocity
Spam is spam, even if it is a genuine email from a reliable company informing me of some truly valuable opportunity... _if I didn't ask the company to send me those emails._
"How to Do Nothing," kids activities, back in print!
It basically comes down to this: Don't click links in email.
People on Slashdot have been doing this for years. That is why nobody reads TFA.
Don't fight for your country, if your country does not fight for you.
Three words too long.
Media that can be recorded and distributed can be recorded and distributed.
-kfg
It's pretty easy to tell the phish from the non-phish, as I don't bank or shop at most of the places the phishers send my way. Also, should I receive an e-mail from my bank (which they already said they wouldn't send me--believing that snail mail is more secure and less likely to be abused), and I feel the need to get there to deal with whatever the message may be saying, I'm surely not going to click a link. Heck, I probably wouldn't even visit the bank during the same session for fear of some kind of redirect spyware that they tried to sneak into the session.
Looking at the URL and seeing "ebay.somewhere.ch" instead of "ebay.com" isn't secure enough anyway, as it's trivial to spoof the status bar with the hover-over text.
The only way to avoid being phished is to not trust any e-mail that has anything to do with anything related to money, savings, charge cards, or deals that are too good to be true--they are too good to be true. A good runner-up is to find a black-hole mail service (i.e., get your own domain name) and set up an account for each vendor you deal with, with a less-than-likely phishable address (e.g. nvrSp4mMy-ebay@mydomain.us). Then, never give your "real" e-mail address to any site you don't explicitly trust. Or even use the same black-hole method for sites you do trust--like slashdot@mydomain.us), instead opting for a black hole e-mail address; this also helps identify who compromised your identity.
While some software is sometimes better at recognizing these things than others (I seldom get phish-mail at my GMail account, as they're recognized and flagged by the other users first), we still can't rely on an automated method to stop these things. It is on the individual to be responsible with their own information.
"I am not who I seem to be," is the safest way to present yourself to the generally anonymous Internet. That's the way they're presenting themselves.
End the FUD
'You know how dumb the average person is? Half of 'em are dumber than that.' Remember, just using computers does not mean someone's got a brain. You only have to work in tech support read some of the many internet message boards to realize that.
Get the picture? Jack of all trade, master of none. Or so goes the old saying. Most of us are good at something. Some could even be called brilliant. I've even met a few people who are very good a most things. I've not yet met one who is good at everything. Not one. I've heard what some of them call some very smart IT people behind their back as well. They call some of *us* idiots because of how well we understand *their* fields.
Seems to me the ones who make it biggest in the IT sector, will be the ones who understand this and can help the people who don't understand computers the best. But then again, those are usually the ones who understand what ROI is and how it affects their jobs, and can actually tell the boss/client why the proposed project should *not* be done. The ones who understand that the person who fell victim to phishing speaks a whole new language that most computer geeks don't understand, just like we speak one they don't.
I expect that this is not a recent phenomenon, nor is it going away anytime soon. Con artists have been around for a very long time. I make the humble sugestion that you vent in here, but for your own sake, please please please don't take it into the work place. It's extremely dangerous to yourself. When perceived as having a negative attitude, most people don't make it far.
funarcadeonline.com
To really be safe, I always call the FDIC before each online transaction to make sure the "bank" I've been dealing with for years even exists.
Don't blame me; I'm never given mod points.
The problem is that while con men target idiots directly like snipers, phishers and spammers pull out a machine gun and mow down everyone on the street.
You might be smart enough not to lose your shirt to a con artist, but if a new one knocks on your door every five minutes, you're going to be pretty damn annoyed.
In the general, low-tech phishing scheme, though, you've just received an e-mail that looks like its legitimately from an organization with whom you do business, and they hope to steal your login and password, or name and SSN by directing you to a look-alike web site, which will give you a "password failed" message. Too late for you now.
I was outlining the not-gonna-happen scenario where one might believe an unwanted/unsolicited e-mail from what looks like an actual bank/other vendor and try to act on that information.
I personally feel a little safer not using Windows, which is the general target of most phishing, or IE when I do have to use Windows. I also have 4 PCs on my KVM, and would most likely follow-up on a separate system entirely...
End the FUD
"That might be a little harsh. We're seeing increasingly sophisticated phishing stuff -- right down to building a look-alike site of the bank which they are pretending to be."
There is absolutely nothing sophisticated about phishing. It is rudimentary at best, and 100% avoidable.
1) If you get business-looking email from someone you don't have an existing business relationship with, it's not legitimate.
2) If you get email with a link to a site you have a business relation with, then type in the URL from the paperwork you got when that business relationship originated; or create a bookmark with the URL manually entered, and use that bookmark to go to the site (all bets are off, though, if you're using Microsoft Internet Explorer). This is especially crucial for banking.
3) If in doubt, use the contact information on the original business paperwork to discuss business.
That's it. You are now phishing free. The sharp-eyed among you may have recognized that these steps are no different than those used for postal mail. These types of scams have been in operation since the dawn of commerce. The only thing that has changed is the delivery mechanism.
I am astonished that people abandon their common sense at the modem (this isn't aimed at the poster. It's just a general observation made at a convenient moment).
I don't think its people abandoning their common sense as you say.
I think that if someone forged a letter which appeared to be from the actual bank you deal with, sent it to you in what appears to be their stationary and envelopes, and used a large amount of legitimate information to indicate that a new department needs to contact you and gave you a corresponding 800 number to call --- you could well fall for it. That's not far from the sophistication being shown nowadays by phishers.
Since its not uncommon for a single entity to use a bunch of annoying domains and email addresses (my ISP/cable company has sent me their marketing crap from a completely innumerable number of e-mail addresses and marketing-driven domains), it's completely possible to get swindled by these.
Yes, for the highly paranoid and tech savvy, we're probably pretty unlikely to fall for this. But, in general, I see a lot of evidence that the level of sophistication behind this is growing -- to the point that even those of s who are tech savvy and paranoid might be hard-pressed to be really sure.
In case you haven't noticed, businesses seem to create new domains for promotional purposes/customer contact all of the time. I know because I keep updating the spam filters for the shite my cable company/ISP sends to me that I don't want to see. The e-mail and or domain is always a permutation of the company name and why they're bugging me. But, it's purely the marketing weenies doing this on behalf of legitimate companies that make this more difficult.
I figure if I can't possibly write enough spam filters to always block them out, lots of people might take slight variances in stride -- because companies make these sorts of changes all the bloody time, thereby conditioning users to expect it to happen. Thereby invalidating much of the advice for avoiding phishing.
Lost at C:>. Found at C.
There used to be a test; back before connecting to the Internet was a matter of plugging the cable from your cablemodem into the back of your computer and clicking 'OK' on all the prompts, you actually had to have enough technical savvy to be able to set up your own TCP/IP stack; even for basic dialup shell access (pre-GUI), you needed to be able to figure out Unix command-line functions. This meant that the people who were posting to the newsgroups were almost always people who had exhibited a minimum level of technical skill. The exceptions were freshmen at college getting access to the Net through their institution's terminal farms, and who could readily be identified by the wave of "Greetings. My name is David Rhodes..." pyramid-scheme postings that heralded the start of each semester and trickled off as they had a little common sense mailbombed into them (if only 0.1% of the readers of a newsgroup emailed someone with an explanation of why it's a pyramid scheme, it still floods their mailboxes).
However, as time went on, the various online services (Delphi, GEnie, et al.) began to offer access to the Net as another feature of their service, with their install software being automated, so if you could stick an AOL floppy into your computer, you could get Net access. And with each new online service that added Net access to their services, you saw a flood of people being exposed to the chain letters and pyramid schemes that had maintained a hand-to-mouth existence on the twice-yearly crop of gullible freshmen -- and there was a steady stream of fresh meat arriving as more people subscribed. With the massive expansion of potential victims, it became a lot more profitable to run scams, and the 'market' boomed, with increased automation making it just as easy to spam the world with 'opportunities' as it was to filter newsgroup postings to find accounts that hadn't posted before and spam them directly.
Based on your numbers, that would mean that 95% of the people on the internet should not be on the internet to begin with. There should really be some training required before you can run an internet connection to your computer.
Similarly, nobody should be allowed to drive unless they can name every part of the car and explain what its function is. Then, they shouldn't be allowed to take a car out onto any roads until they've studied civil engineering.
Really, the idea that everyone who uses a service should know what's going on inside the black box is just stupid.
"The use-mention distinction" is not "enforced here."