The biggest scare I've gotten is somehow ending up on "colnbase.com" (instead of "coinbase.com").
It's defunct now, but at the time it was a 1:1 replica of Coinbase. And the only reason I noticed was because 1Password didn't offer to fill in my credentials.
While knowing someone's email/password combo might not be enough for an attacker to do anything malicious on Coinbase itself (due to email re-verification maybe), the point is that even the smartest of us Hacker News users can fall for it. And that should scare the rest of us.
So so true. 1Password refusing to auto fill a password has saved me multiple times in the past! Also, one of my friends has a PhD in literally rocket science (aeronautical engineering from MIT) and got scammed by someone who stole his brother's SIM card and did some shenanigans. No one is safe, no matter how smart or tech savvy you think you are! For the less tech savvy folks, I understand why they are scared, it's hard to give them even general tips to not lose the farm to fraudsters.
Don't trust incoming calls, text messages or emails.
Don't trust caller ID on your phone.
If someone calls you asking for information or to do something, ask for a case id or reference number. Hang up, call back on a number you get from a previous bill, back of your credit card, or by googling the company.
If anyone is pushing for something to be done urgently, stop. Hang up, don't take any action. Call a trusted other person and talk to them about it.
I know that "rocket scientist" has been a standin for "smart person" or "genius", but in this case, I would be more surprised if a computer security expert (various job titles) had been scammed, because it's their job to be up on this stuff.
How often does a rocket scientist deal with computer viruses, or phishing emails, etc compared to a security expert? Most of the time, their IT security expert (ideally) stops it before it gets to them..
> Also, one of my friends has a PhD in literally rocket science (aeronautical engineering from MIT)
One of my friends is a nuclear physicist from TU Delft and they somehow managed to install a fake clone of Chrome haha. Somehow never got their accounts broken into or money stolen.
Physical SIM cards should _always_ have a strong PIN set. It baffles me how many people either don't have a PIN or it's just set to 0000. You're basically handing over your whole digital life away to anyone who gets their hand on your SIM card.
I nearly lost an account because I assumed that 1Password was just being dumb not offering to auto-fill credentials. Turns out I'm the dumb one for doubting it.
Now if 1Password shows nothing to auto-fill I make damn sure I'm on the right site.
> the only reason I noticed was because 1Password didn't offer to fill in my credentials.
Nice, I always hope this will save me but I never landed on such a phishing site. How did it happen for you?
About domain-based autofills, perhaps less so now than 5-10 years ago: it always seemed weird that the whole security industry seemed to say these plugins, or the browser's built-in password store, are dangerous because there were past vulnerabilities and any website you visit can exploit it. The way I see it: vulns get fixed, I just need to not be in the 1st wave of persons they target (risk type: plane crash, very small odds but sucks to be you); receiving phishing emails or messages happens constantly and apparently it works well enough to continue doing it and evading filters constantly (risk type: car crash, can happen and they get only the creds for the website being autofilled). Would recommend to anyone who then realises something is up when the autofill doesn't work, but ideally would have more evidence to back that up
One of the worst parts of using a oassword manager is that apps and websites don’t by default share their credentials. I could totally see me getting caught by a shady link to a website of an app that I use but because I’ve never logged into the website, 1Password makes me search for it.
My friend was not smart as you, and religiously typed password on a fake Amazon website link clicked from an SMS promising a refund on recent purchase. Stopped only when it asked for 2FA code because there was no 2FA setup.
> The biggest scare I've gotten is somehow ending up on "colnbase.com" (instead of "coinbase.com").
You might want to install some browser extensions to block content. Then block all content (set to whitelist) and selectively add the sites you know.
If you end up on a new site with some amalgamation of letters that look familiar, the extension will rightfully block it and prompt you whether you want to whitelist or not. Big ole' red flag right there.
Of course it's not foolproof. It is just another layer in the strategy of defense-in-depth.
As usual this started with an incoming phone call. If you ever receive a phone call from a tech company, it's a scam. The caller ID doesn't matter. The caller's accent (wtf) doesn't matter either. It's a scam.
If the consequences for letting that call go to voicemail are any less severe than full account takeover by a script kiddie, you're still better off never picking up.
Google in particular is famous for making it impossible to contact a human. If Google calls you, before picking up, consider whether you truly believe you're lucky enough to be one of a handful of people in the world to ever get human support from them.
You still always assume an incoming call is a scam no matter what. Hang up, look up, call back, in that order.
Very occasionally you might be making some poor customer support person's job harder, but the vast majority of the time you'll be hanging up on a scammer. You can be polite about it, but firm and brief. "It's my policy to always call back no matter what, nothing personal."
I do this for any inbound call, unless the caller id is someone in my contacts it can go to voicemail. If it's important they either leave a voicemail or keep trying, for repeated calls I will answer but with skepticism.
If they are in my contacts I will recognise their voice.
I regularly get phone calls from Google because I helped a friend with their ad account once. No matter how many times I tell them not to call they eventually find a new number to call me from. They are legit calls. Google just won't take no for an answer.
This is the same type of phishing attack described here[1]. It’s still surprising to me how the SPF, DKIM, and DMARC all pass. If I remember correctly, it’s because they actually have a clever way od getting Google to send an email to you by sharing a Google Form with you or something like that.
> Hack Clubbers have determined that this is almost definitely a bug in Google Workspace where you can create a new Workspace with any g.co subdomain and get it to send some emails without verifying that you own the domain.
Seems like this is the flow:
1. Create a Google Workspace with a g.co subdomain. Apparently this is not verified, or verifying the domain is not necessary for the next steps.
2. Create an account for the victim under this Google Workspace.
3. Reset that account's password.
The victim gets an email from Google Workspace informing them that their password was reset. And this email is a real, legitimate (not spoofed) email from Google because it's just a result of the normal password reset process for a Google Workspace account.
We have all this verification on the Web, but not the phone. Why do telcos allow for spoofing? We wouldn't allow that with email. Is this a technical limitation that allows for spoofing?
We have had phone company employees here explaining that their company makes a lot of money from scammers/spammers/etc so they do not have an incentive to stop it.
What's even more interesting is there is no DNS records for important.g.co, which means they have found a way to create an Google Workspace without verifying the domain but still able to send emails like password resets.
It's definitely a glitch where you can send emails/transactional emails from an unverified Google Workspace. My guess is that their are protections for google.com and google domains but they forgot to add the g.co domain, which allows unverified sending to g.co and creation of workspaces.
I'm not sure if it's good thing or not but I've come to consider that any notification about a password being reset or a fraudulent charge is phishing unless I initiate some action.
I always verify that I'm actually fucked and then take action. This seems counter-intuitive but the deluge of phishing emails makes me feel this is the safest option. I'd rather wait to notice a fraudulent charge and dispute it, than leak info to a random SMS number that claims (possibly truthfully) that someone in Japan spent $9000 at the gucci store.
Agreed. I do not follow any links, accept calls, etc. I go to the site of origin and do what I need. Also be careful if you search for the sites name on Google, still might click a fraud site!
That's not verifying the phone number. I received a call from Chase about a wire. I asked them for a code so I could continue the conversation and then looked up the phone number on their website and called that and talked through reps till I got to the right department.
Caller ID being spoofed is the wrong way to think about this. It's just that if someone walks up to you and says "Hey, I'm Jean d'Eau and I'm President of the US" you don't think to yourself "oh yeah he's definitely President and that's his name".
People can always tell you they're whoever they want to be. You can either believe it or go find out if they are.
I know it's easy to second-guess someone after they've explained that they're describing a scam, but:
> The thing that's crazy is that if I followed the 2 "best practices" of verifying the phone number + getting them to send an email to you from a legit domain, I would have been compromised.
He didn't follow the first of those best practices. He just looked up a phone number that the caller also read out to him, and didn't call it. And "Solomon" also explicitly told him he couldn't call.
I honestly think that at this point, no incoming phone call can ever be trusted.
I don't even know where the idea that those are the best practices came from.
The phone number best practice has always been constructed as "call them back at a known good number, preferably one written on paper or on your card". You certainly don't ask them to show you where on the company website the phone number is listed.
And asking the person on the phone with you to send you an email from a specific domain is likewise not something I've ever seen recommended: that's one of several things you check to see if an email is phishing (And only one of several! A good domain isn't enough to clear an email!) But if you're already on the phone with someone suspicious, the best practice has always been to get off the phone with them immediately and call a known number, not to ask the caller to prove themselves.
None of this is to blame OP for misunderstanding, it's just very clear that we need to do better at communicating these rules out to the world.
They can't. And they haven't been for a while. Spoofing phone calls is simply too easy, and nothing is being done to fix that, despite the fact that it puts so many of us at risk. It's not an insurmountable problem, technologically. It is literally a lack of will and outcry from ordinary people, despite how often this fact is used to abuse so many.
Credit Card companies have known this for a long time. My credit card company will call and say "do not call back to this number, call the number on the back of your card and use this reference number".
Telecoms know if a number is spoofed or not. All I want is for them to wholesale steal the original Twitter "verified" check, and use it to confirm that a call is not spoofed.
I'd argue the second one was not followed either. Maybe I'm misunderstanding the article, but I would not take a random "your password has changed" as proof. I would need the caller to send me an actual email from their personal work email address (or ticket system?) with some actual, human communications in it.
Readit News
It's defunct now, but at the time it was a 1:1 replica of Coinbase. And the only reason I noticed was because 1Password didn't offer to fill in my credentials.
While knowing someone's email/password combo might not be enough for an attacker to do anything malicious on Coinbase itself (due to email re-verification maybe), the point is that even the smartest of us Hacker News users can fall for it. And that should scare the rest of us.
Don't trust incoming calls, text messages or emails.
Don't trust caller ID on your phone.
If someone calls you asking for information or to do something, ask for a case id or reference number. Hang up, call back on a number you get from a previous bill, back of your credit card, or by googling the company.
If anyone is pushing for something to be done urgently, stop. Hang up, don't take any action. Call a trusted other person and talk to them about it.
How often does a rocket scientist deal with computer viruses, or phishing emails, etc compared to a security expert? Most of the time, their IT security expert (ideally) stops it before it gets to them..
One of my friends is a nuclear physicist from TU Delft and they somehow managed to install a fake clone of Chrome haha. Somehow never got their accounts broken into or money stolen.
This is one of the reasons scammers like to target doctors.
Now if 1Password shows nothing to auto-fill I make damn sure I'm on the right site.
Nice, I always hope this will save me but I never landed on such a phishing site. How did it happen for you?
About domain-based autofills, perhaps less so now than 5-10 years ago: it always seemed weird that the whole security industry seemed to say these plugins, or the browser's built-in password store, are dangerous because there were past vulnerabilities and any website you visit can exploit it. The way I see it: vulns get fixed, I just need to not be in the 1st wave of persons they target (risk type: plane crash, very small odds but sucks to be you); receiving phishing emails or messages happens constantly and apparently it works well enough to continue doing it and evading filters constantly (risk type: car crash, can happen and they get only the creds for the website being autofilled). Would recommend to anyone who then realises something is up when the autofill doesn't work, but ideally would have more evidence to back that up
Just copy one of Coinbase's legit emails for something like "A withdrawal of $1,200 USD has been started" and you have the perfect bait.
Well, ok then.
But you didn't fall for it, a simple password manager technique worked as advertised?
There’s indeed a lot of them :)
You might want to install some browser extensions to block content. Then block all content (set to whitelist) and selectively add the sites you know.
If you end up on a new site with some amalgamation of letters that look familiar, the extension will rightfully block it and prompt you whether you want to whitelist or not. Big ole' red flag right there.
Of course it's not foolproof. It is just another layer in the strategy of defense-in-depth.
Google in particular is famous for making it impossible to contact a human. If Google calls you, before picking up, consider whether you truly believe you're lucky enough to be one of a handful of people in the world to ever get human support from them.
Very occasionally you might be making some poor customer support person's job harder, but the vast majority of the time you'll be hanging up on a scammer. You can be polite about it, but firm and brief. "It's my policy to always call back no matter what, nothing personal."
You don't have to pretend to be confused.
The industry of Indian scam call centers is not a crazy conspiracy invented by racists.
Nor was the industry of Indian legitimate call centers.
You cannot glean any useful signal of legitimacy from the caller's accent.
That's the WTF.
Deleted Comment
If they are in my contacts I will recognise their voice.
[1]: https://news.ycombinator.com/item?id=42450221
> Hack Clubbers have determined that this is almost definitely a bug in Google Workspace where you can create a new Workspace with any g.co subdomain and get it to send some emails without verifying that you own the domain.
Seems like this is the flow:
1. Create a Google Workspace with a g.co subdomain. Apparently this is not verified, or verifying the domain is not necessary for the next steps.
2. Create an account for the victim under this Google Workspace.
3. Reset that account's password.
The victim gets an email from Google Workspace informing them that their password was reset. And this email is a real, legitimate (not spoofed) email from Google because it's just a result of the normal password reset process for a Google Workspace account.
How about a law with teeth?
We allowed email to be the wild west for years and I'm not sure it's better than telephony now
It's definitely a glitch where you can send emails/transactional emails from an unverified Google Workspace. My guess is that their are protections for google.com and google domains but they forgot to add the g.co domain, which allows unverified sending to g.co and creation of workspaces.
I always verify that I'm actually fucked and then take action. This seems counter-intuitive but the deluge of phishing emails makes me feel this is the safest option. I'd rather wait to notice a fraudulent charge and dispute it, than leak info to a random SMS number that claims (possibly truthfully) that someone in Japan spent $9000 at the gucci store.
Caller ID being spoofed is the wrong way to think about this. It's just that if someone walks up to you and says "Hey, I'm Jean d'Eau and I'm President of the US" you don't think to yourself "oh yeah he's definitely President and that's his name".
People can always tell you they're whoever they want to be. You can either believe it or go find out if they are.
> The thing that's crazy is that if I followed the 2 "best practices" of verifying the phone number + getting them to send an email to you from a legit domain, I would have been compromised.
He didn't follow the first of those best practices. He just looked up a phone number that the caller also read out to him, and didn't call it. And "Solomon" also explicitly told him he couldn't call.
I honestly think that at this point, no incoming phone call can ever be trusted.
The phone number best practice has always been constructed as "call them back at a known good number, preferably one written on paper or on your card". You certainly don't ask them to show you where on the company website the phone number is listed.
And asking the person on the phone with you to send you an email from a specific domain is likewise not something I've ever seen recommended: that's one of several things you check to see if an email is phishing (And only one of several! A good domain isn't enough to clear an email!) But if you're already on the phone with someone suspicious, the best practice has always been to get off the phone with them immediately and call a known number, not to ask the caller to prove themselves.
None of this is to blame OP for misunderstanding, it's just very clear that we need to do better at communicating these rules out to the world.
But you're right: simply say "given that this is a sensitive security matter, thank you for the heads up. Don't call me, I'll call you (click)"
They can't. And they haven't been for a while. Spoofing phone calls is simply too easy, and nothing is being done to fix that, despite the fact that it puts so many of us at risk. It's not an insurmountable problem, technologically. It is literally a lack of will and outcry from ordinary people, despite how often this fact is used to abuse so many.
Credit Card companies have known this for a long time. My credit card company will call and say "do not call back to this number, call the number on the back of your card and use this reference number".
That should absolutely be the norm at this point.