
UPS delivers some smishing advice (but have they kept something under wraps?), we ask ChatGPT to take a long hard look at itself, and we debate what the penalty should be for taking national secrets home with you.
All this and much much more is discussed in the latest edition of the “Smashing Security” podcast by cybersecurity veterans Graham Cluley and Carole Theriault, joined this week by Host Unknown’s sole founder Thom Langford.
Warning: This podcast may contain nuts, adult themes, and rude language.
Show full transcript ▼
This transcript was generated automatically, probably contains mistakes, and has not been manually verified.
Hello, hello, and welcome to Smashing Security, episode 328. My name's Graham Cluley.
I can't go, but I'm sure Thom Langford will pick them up for us.
She wants to make sure that I'm there so I can pick up. Yeah. Because I had three things in the mix. I, you know, statistically I was— Oh dear. But no, couldn't believe it.
Most entertaining cybersecurity podcast and best all-rounder cybersecurity podcast or something like that.
But before we kick off, let's thank this week's wonderful sponsors, Bitdefender, Collide, and Drata. It's their support that help us give you this show for free.
Now, coming up on today's show, Graham, what do you got?
And I think we actually need to praise companies sometimes when they raise awareness as to the threats which are out there and give a little bit of it.
So I thought I'd do something a little bit different.
Anyway, listen, I thought, let's actually applaud a company doing something right, because UPS in Canada, the delivery firm, has gone out of its way to contact customers.
They sent them a letter, and I thought it's worth reading out because there's some great advice in here, which I think would be suitable for everyone who listens to the show.
So you get this letter and it says at the top, 'Fighting phishing and smishing: an update from UPS.' Okay, that's all right, isn't it? 'At UPS, we are committed to fighting fraud.
We want to let you know what phishing and smishing are, and what you can do to protect yourself.'
And now it seems people are saying spearfishing when there's an attachment, whereas I always view phishing as something—
Fraudulent emails referred to as phishing and text messages, referred to as Smashing Security, are becoming more common. That's true.
Fraudsters attempt to convince package recipients that they owe money for delivery of a package and send text messages or emails to solicit credit card and other payment card data.
I mean, we've all seen that, haven't we?
And then if I get a text and I know something's coming, I'm like, did I get everything? Am I waiting for something? Is this one? Just so my husband, he goes, no, fuck off.
Well, I— no, because they have your address, but they don't have your mobile phone number normally, do they? I don't know, that's the sneaky thing.
These messages may appear legitimate by incorporating company brands, colours, or other legal disclaimers.
'These fraud attempts affect deliveries from many carriers.' Brackets, in other words, not just UPS.
You can learn more about common types of fraud and see examples of fraudulent messages at incredibly long URL, right? Okay, the letter goes on. 'Have you been smished?
If you've received something that doesn't look or feel right, trust your instincts. Real UPS texts,' at least in Canada, 'will only come from SMS number 69877 in Canada.'
UPS has been working with partners in the delivery chain to try to understand how that fraud was being perpetrated.
As part of that effort, UPS conducted an internal review to assess whether information it received from shippers was contributing to this fraudulent conduct.
In other words, is some information leaking out?
Well, the next sentence, Carole: During that review, UPS discovered a method by which a person who searched for a particular package or misused a package lookup tool could obtain more information about the delivery, including a recipient's phone number.
In other words, they've snuck in, and we're in about paragraph 5 or 6 now. We might have had an issue.
They're just telling me what phishing and smishing are.' Yeah, you wouldn't have got past the first paragraph.
You may have been impacted by this." So they're saying their package lookup tool has been leaking recipients' names, shipment addresses, potentially phone numbers, order numbers.
It says, "We can't tell you exactly when this has been happening, but it looks like it has been happening to some customers from February 2022 until the end of April 2023." Blimey!
So those texts, if you'd received one, it may have been a lot more convincing because—and this is thanks to the folks at Bleeping Computer—they've uncovered people who were expecting deliveries from UPS who got very, very convincing messages.
Right. Now, Thom, you are a big fan of Apple tech, and you're also a huge fan of LEGO, aren't you? Yeah.
Well, we're going to link in the show notes to a couple of examples from people who were expecting deliveries via UPS of a LEGO order.
And they got text messages saying, "Your LEGO order is waiting delivery to your shipping address, postal code blah blah blah.
You need to pay a shipping fee in order to have the parcel on time. To avoid delays, click here."
What they want to get are the people who—the more gullible ones—maybe don't—yeah, who maybe don't quite have the same sort of, you know, cognitive abilities to see.
It's not just cognitive, it's digital ability, right?
When you go to the URL, it's going to grab other personal information or charge your card more than that.
So there are all manner of potentially, you know, people who are falling for much more convincing delivery failure, or "you need to act upon this UPS message," smishing cam—I hate the word smishing—campaigns than ever before.
What's the advice? The advice? Don't call us, we'll call you.
They've hidden it as much as possible behind what looks like a generic piece of—
In fact, I found a Wikipedia page that lists all of the famous whistleblowers going all the way back to the 1600s, which is a rabbit hole you don't want to go down.
Did it involve a rabbit?
And so, yeah, so it has been going back at least 3,000 years. Exactly.
It's from the Office of Public Affairs of the US Department of Justice, and it talks about a former FBI analyst who was sentenced for retaining classified documents.
So there's this FBI analyst, her name is Kendra Kingsbury, 50, of Garden City, Kansas, and she was sentenced to 46 months federal prison followed by 3 years of supervised release.
And the reason for that is, she pled guilty to 2 counts of unlawfully retaining documents related to national defense.
Bottom line was, she was an analyst for the FBI for 12 years, and she was caught taking a whole bunch of confidential documents away and taking them home, basically.
Now, she held a top-secret security clearance, so she could see confidential, secret, and top-secret documents.
All of the documents she took were classified as secret, so the middle level, but many of which include documents that describe intelligence sources and methods related to US governments, all to do with counterterrorism, counterintelligence.
Also included in numerous documents classified as secret from other government agencies— Oh boy.
—describing intelligence sources related to US government efforts to collect intelligence on terrorists.
Well, the investigation—and this isn't even the crux of it—the investigation actually turned out more questions than answers, because when they analyzed and reviewed her telephone records, revealed a number of suspicious calls, including numbers associated with subjects counter-terrorism investigations.
And those individuals also made calls back to Kingsbury. So there's obviously something going on here, right?
You know, so not only did she take these documents where she wasn't supposed to, all classified at secret level, not top secret level, but there was subsequently found to be some kind of sharing of said documents and other activity.
Now that took me down another rabbit hole because as I said, she was sentenced to, what was it, 46 months.
That took me down the rabbit hole of a woman called Reality Winner, which is not the name of a TV show on Channel 5, but she was an analyst in the NSA. She was a translator there.
She released one document to the press, which was basically information about Russian interference in the 2016 election. She was arrested, obviously.
I mean, you found this stuff has been released, et cetera—not good, although you could say it's for the greater good.
She was charged with removing classified material from a government facility and made it to a news outlet.
She was denied bail and then sentenced to 63 months in prison, which if you do the sums, 5 and a half years in prison.
So for releasing one document compared to this other person, Kingsbury, who stole a whole bunch of documents, made some dodgy phone calls, sentenced to 4 months.
But what happened was, she printed out some of this sensitive information at her workplace.
And she gave those printouts to reporters at The Intercept, which was the news outlet who reported it.
This is useful information, by the way, if you're planning to write a ransom note or something like that. Your printer.
And so here's two cases, just two cases where we see secret documents, even a top secret document, potentially being leaked by a servant or—
So bottom line is we've got Donald Trump has taken boxes of material, said he's returned all of it. He absolutely hadn't, despite very clearly stating that he had.
When his Mar-a-Lago residence was raided, there was stuff found everywhere, in public places, in a ballroom and a bathroom and all that sort of stuff, top secret documents allegedly relating to nuclear secrets and stuff like that.
And not only was Donald Trump not requested to post bail, he's certainly not been arrested and is basically throwing money at the problem to try and make it go away.
I'll store it in the highly secure loos at Mar-a-Lago.
People won't expect the highly sensitive information to have been left accessible to anyone.
And I just, I just find this utterly amazing how—
It's not a case of if you can't do the time, do the crime. It's just more a case of if you can afford to do the crime, then crack on because nobody's going to catch you.
I talked to lots of people. I met a cool chick on a plane. I met a great chef.
I met an Airbnb host who thought that people staying in a non-air-conned pad would rejoice at pure, 100% pure polyester sheets. So that was really fun.
You should have seen my Yeti of a husband.
Like overlooking the rolling hills of Istria, right? Like, think super poshy posh, mismatched fabrics and pop art and terrazzo floors and big, big lights, okay? Like the whole thing.
And we just were going there for just a, you know, a Coke to look at the sunset. But I'm gonna send you the art that was outside the front.
I'm gonna send it on our little text message thingy here. Okay? So take a look and maybe one of you—
But I think it's 'cause they don't want kids there. I think it's an adult hotel, that kind of thing. It's not a family hotel. So maybe these are just to scare off the kids.
And some would go and look at art and some would listen to the pods.
And one of them called me up afterwards and said, look, I've just listened to 3 episodes of Smashing Security in a row. And you guys are amazing. You're great. You're wonderful.
It's all tech speak. You know, you talked about ChatGPT or whatever, and I couldn't follow, right? And this lady is a GP. Oh, you see GP, GPT.
But you know, she's brainy, she's funny, but our stupid tech-only lingo kind of puts up this anti-learning fence.
So I am sorry to her and all the other listeners, and I'm going to try and describe it here in a way, but there's a good piece of info for you guys that know this inside out at the end.
So stay with us and you guys are going to help me. Okay. If I say something too techie, you just go, let me just describe what it is.
But she has liked the recent trends of having Mr. Cluley on because she really likes Graham's voice. She finds it very, very— oh, how lovely. Very warming.
And so what the heck is it? Well, I thought, why not ask ChatGPT? Right? It said ChatGPT is an advanced conversational AI model developed by a company called OpenAI. What? What? AI? AI?
Sorry, what's AI? Artificial intelligence. Very good, Thom. Thank you. I didn't spot that one. I was listening. Very good.
Number 2, ChatGPT is trained on a diverse range of internet text sources to learn patterns, grammar, and context in order to generate coherent and contextually appropriate responses.
Now, apparently the dataset has at least 300 billion words in it. So diverse, I think, is a little misleading here.
I think, you know, gluts and gluts and gluts of stuff that they could find is maybe perhaps more realistic. Would you guys agree?
So basically, but the thing is, it's a tool right now available to anyone that speaks the supported languages, I guess, right? Anyone with internet access.
What you can do is go to openai.com and you will find ChatGPT there, right? It's free to use. But you have to create an account and there's nothing to learn or set up.
Basically a search box like any search engine, and you can put in a question and allons-y, right? You see what crops up.
So you could ask a question about medicine or real estate or mythical monsters or recipes or help me out, poetry.
Which is why investors are tripping over themselves to get on the AI— sorry, artificial intelligence model train. Choo choo all the way to the bank.
Now, the thing is that there is a catch, right? You cannot trust the information spouted by ChatGPT to be 100% correct. Any of the time, I would say.
When Mark Stockley was on a few weeks ago, he was telling us about those, that law case where ChatGPT was coming up with fake past verdicts.
Fake cases and, you know, and it was persisting in claiming that these things were real and they weren't and it was just making it up.
So in short, ChatGPT's mama is the internet and it gorged, okay, I'm gonna say it, at the internet mama nipple until it was ready to be unveiled to the world.
And you can go look at our backlog of Smashing Security episodes because we've talked about it a lot.
And the question is, is who decided to allow ChatGPT or any of these artificial intelligence models into the public world? So I thought, I'll ask ChatGPT.
And it said it was made by the organization or company responsible for the development and deployment. In this case, OpenAI and ChatGPT, the decision was made by OpenAI itself.
And the point I'm making is there's no regulatory oversight here. It's just one company going, okay, we're ready. Are we ready? Let's go.
I mean, imagine how much it would constrict Thom, for starters, with what he gets up to on the internet.
Everything that you can find on the internet freely is there freely, right? It could have just gone to a library.
If it had a body and fingers and eyes, it could have gone to libraries and read everything in a library, you know?
I looked at what jobs are at high risk with ChatGPT on the horizon.
The jobs that are— there's Morgan, Piers Morgan— jobs that are deemed most safe include athletes, car repair people, cooks, and get this, this is my favorite, stonemasons.
Stonemasons, you guys are fine. So high five to you for, you know, having not gotten on the digital bandwagon. Well done.
I mean, it's not the most, you know, common of requirements. They're not gonna be creating Gothic arches for these massive cathedrals of battery power.
Currently, I think, correct me if I'm wrong, the winner in the front is OpenAI at the moment, right? They have the lead.
But yesterday, Google's DeepMind CEO, mic drop, that his new AI algorithm, soon to be on the digital shelves, will eclipse ChatGPT. Oh, for goodness sake.
Okay, so if you're interested in trying out ChatGPT and you don't know what it is and you've heard people talk about it, do not go to Facebook or social media and click on a 'Try ChatGPT' ad.
No, no, no.
Okay, so Smashing Security company in Purva said that they saw some scams pretending to be access to these, you know, AI models, artificial intelligence models and the like.
So just use your web browser and go to openai.com.
And second tip, if you decide to use ChatGPT, know that your questions are logged by default and some people keep sessions going tied to your account because you need to have a user login to get in to use it now.
So to change this, once you've created an account, you can click on your username to the settings and clear all chats.
And you can also go to the data controls and disable chat history and training.
That's a great point, actually, Carole, because I mean, a company— some companies are blocking access to ChatGPT because it just produces garbage sometimes and low-quality content.
But the more serious point is that people are feeding in sensitive information into ChatGPT, which is then being collated and used, and maybe company sensitive, and maybe other people's personal details, all sorts of things.
Exactly.
And they think that's exactly the reason that people are using it and kind of feeding in sensitive information without realizing it.
And those logs are super delicious to someone who might want to try and attack the company.
So there you go, and don't use an easy-to-guess password if you're going to create a login on ChatGPT, okay? Try something that's unique and impossible to remember.
It'll probably search the internet to find out the list of the top passwords, and it'll think, oh, that's number one, let me use that one. I thought it was password1. It probably is.
With an exclamation mark afterwards.
So, if we were to say one word a second and if we were then to say 300 billion words, how many years do you think that would take us to complete?
With over 14 frameworks including SOC 2, GDPR, HIPAA, and ISO 27001, Drata gets you audit-ready for crucial security standards needed to scale your business.
Automated controls, over 75 integrations, and 24-hour monitoring keeps your company in compliance without manual work.
And with a new open API and plenty of customization, you can build your program your way. With over 360 5-star reviews, Drata is the highest-rated cloud compliance platform on G2.
Countless security professionals from companies Notion Lemonade and Bamboo HR have shared how crucial it's been to have Drata as their trusted compliance partner.
So listeners of Smashing Security, you can get 10% off Drata and waived implementation fees at smashingsecurity.com/drata. That's smashingsecurity.com D-R-A-T-A.
If a device isn't compliant, the user can't log into your cloud apps until they fix the problem. It's that simple.
Kolide patches one of the major holes in zero trust architecture: device compliance.
Without Kolide, IT struggles to solve basic problems keeping everyone's OS and browser up to date.
Insecure devices are logging into your company's apps, but there's nothing there to stop them.
Kolide is the only device trust solution that enforces compliance as part of authentication, and it's built to work seamlessly with Okta.
The moment Kolide's agents detect a problem, it alerts the user and gives them instructions to fix it. If they don't fix the problem within a set time, they're blocked.
Kolide's method means fewer support tickets, less frustration, and most importantly, 100% fleet compliance. Want to learn more? Of course you do. Visit kolide.com/smashing.
That's kolide.com/smashing. And thanks to Kolide for sponsoring the show.
Now, did you know that you can log into Bitwarden using a secondary device instead of your master password? Well, now you do.
Logging in with a device is a passwordless approach to authentication.
It removes the need to enter your master password by sending authentication requests to other devices you're currently logged into for approval.
With Login for Device, it can be initiated on the Web Vault, browser extension, desktop app, mobile app, and you can approve access on your mobile and desktop app version of Bitwarden.
Very, very cool. And the Bitwarden team has hardened the security of its vaults, protecting new vaults with 600,000 iterations by default.
And of course, existing accounts can also update themselves to the same level. These and many other great security features are incorporated all the time into Bitwarden.
Keeping your passwords secure from hackers. Learn more, try Bitwarden for yourself at bitwarden.com/smashing. That's bitwarden.com/smashing. And welcome back.
And you join us at our favorite part of the show, the part of the show that we like to call Pick of the Week.
Could be a funny story, a book that they've read, a TV show, a movie, a record, a podcast, a website, or an app, whatever they wish.
It doesn't have to be security-related necessarily. Better not be. Well, my pick of the week this week is not security-related.
My pick of the week this week is a documentary, which I have— I love a documentary, as you know. Again? Yes.
So, my pick of the week this week is a documentary called Speedcubers. Ooh! Which you can find on Netflix.
And it's a documentary all about people who are incredibly good at solving the Rubik's Cube. Didn't take long then, the movie. It's about 40 minutes, the documentary.
His world record attempt is 3.13 seconds. That's his world record. Wow. One-handed, he can do it in 6 seconds, just with one hand.
Anyway, it's a really touching story. Max Park is severely autistic. And doing the cube has helped enormously with his life.
And Feliks Zemdegs from Australia was the guy who Max Park looked up to, and they became great buddies, and then they began competing against each other.
But they have a genuine and lovely friendship, and you kind of think, what a lovely couple of guys. Must be nice.
So I recommend Speedcubers, Netflix documentary, all about the Rubik's Cube and the masters of the cube. I really enjoyed it.
Now, on kbdcraft.com, you can buy mechanical keyboards.
Now, mechanical keyboards, for those who don't know, they're the old-style IBM clacky-clacky keyboards rather than the laptop-style keyboards that we often use now.
And there is a whole subculture of building your own keyboards and customizing it, so the little microswitches underneath, have different pressures and noises and sensitivity and all that sort of stuff.
Absolutely fascinating.
And then it's a fucking keyboard.
But the unique thing about— yes, the KBDcraft website is not only do you get to customize your keyboard, as it were, you actually get to build the entire frame.
So not only do you get the base of your, you know, which you push all the little switches into, they put the keys on top and all that.
You get to build the frame out of Lego, or I should say compatible to Lego.
They offer white and gray. You can add things to it.
There are also instructions because their initial kit is called the ADAM, A-D-A-M, and then they've got a numeric keypad called the Kit Adams.
Took me a while to work out ADD AMS, because that's what you used to add stuff with.
They're all currently wired at the moment, USB-C, but I'm sure that, you know, Bluetooth will be coming along soon at some point.
You can— the keyboard is backlit, and you can download, you know, an open-source app that allows you to customize the keys, and the colours.
And it's not even LEGO, is it? So the case isn't made out of LEGO. It's made out of some generic LEGO rip-off, isn't it?
So you're not loyal to the LEGO Corporation? Oh, I am. I don't buy any other kits. Okay.
This is the first one I've bought that isn't actually LEGO, but then again, LEGO aren't going to make a keyboard and I thought this was quite cool.
It's taken me a little bit of getting used to because I'm not used to proper keyboards.
There's not much travel on it.
I must admit, you know, it's not my favorite go-to type-on thing, but it was really good fun to build, good fun to learn about keyboard mapping and the software behind it and the science behind it.
And it was a nice little construction project. Okay. Okay, okay.
And you follow a computer scientist, you know, a nerdy who becomes obsessed with a mysterious new neighbor.
And you soon find out that the geeky narrator, Nancy, right, uncovers the neighbor's dark secret. She's a centuries-old vampire. See, how fun is that?
And Nancy becomes her familiar, and bringing the vampire into social media, you know, New York City.
And they're both pulled down this huge rabbit hole of deceit and murder and mayhem. So, it's basically, the whole story is vampire versus unhinged stalker neighbor.
What could go wrong? That is basically the premise of the series. It's funny, it's twisty, it's turny, it's a bit gross. They got great sound effects.
I don't know how they did 'em, but I'm sure a big bucket of jelly would help. Cabbages in jelly, probably. It stars Leslie Grace and Sonny Bringas.
It's How to Win Friends and Disappear People. Find it wherever you get your podcasts if you enjoy a good audio drama.
If we can get more listeners commending my documentary suggestions, that'd be great as well. Well, that just about wraps up the show for this week.
Thom, I'm sure lots of our listeners would love to follow you online and find out what you're up to. What's the best way for folks to do that?
So yes, check it out.
And don't forget to ensure you never miss another episode. You can follow Smashing Security in your favorite podcast apps, such as Apple Podcasts and Spotify.
For episode show notes, sponsorship info, guest lists, and the entire back catalog of more than 327 episodes, check out smashingsecurity.com.
What did you say before? Seriously, I'm having a total mind fuck.
But she was wearing, she was wearing this top that just covers her nips, right? So her whole bottom boob is out.
Hosts:
Graham Cluley:
Carole Theriault:
Guest:
Episode links:
- UPS discloses data breach after exposed customer info used in SMS phishing – Bleeping Computer.
- Example of UPS SMS phishing message related to Lego order – Twitter.
- Another example of a Lego-related UPS phishing message – Twitter.
- Former FBI Analyst Sentenced for Retaining Classified Documents – US Department of Justice.
- How The Intercept might have helped unmask Reality Winner to the NSA – Graham Cluley.
- Bad adverts leave people scratching their heads – MSN.
- How Cybercriminals Can Perform Virtual Kidnapping Scams Using AI Voice Cloning Tools and ChatGPT – Trend Micro.
- Which Jobs Will Be Most Impacted by ChatGPT? – Visual Capitalist.
- Unraveling an AI Scam with AI – Imperva.
- 100,000 Hacked ChatGPT Accounts Discovered on Dark Web – Hackread.
- 97+ ChatGPT Statistics & User Numbers In June 2023 (New Data) – Nerdy Nav.
- “Speed Cubers” – Netflix.
- Trailer for “Speed Cubers” – YouTube.
- KBDcraft.
- ”How to Win Friends and Disappear People” – Qcode Podcasts.
- Smashing Security merchandise (t-shirts, mugs, stickers and stuff)
Sponsored by:
- Bitwarden – Password security you can trust. Bitwarden is an open source password manager trusted by millions of individuals, teams, and organizations worldwide for secure password storage and sharing.
- Kolide – Kolide ensures that if your device isn’t secure it can’t access your cloud apps. It’s Zero Trust for Okta. Watch a demo today!
- Drata – With over 14 frameworks including SOC2, GDPR, HIPAA, and ISO 27001, Drata gets you audit-ready for crucial security standards needed to scale your business. As a listener to Smashing Security you can save 10% off Drata and have implementation fees waived.
Support the show:
You can help the podcast by telling your friends and colleagues about “Smashing Security”, and leaving us a review on Apple Podcasts or Podchaser.
Become a supporter via Patreon or Apple Podcasts for ad-free episodes and our early-release feed!
Follow us:
Follow the show on Bluesky at @smashingsecurity.com, or on Mastodon, on the Smashing Security subreddit, or visit our website for more episodes.
Thanks:
Theme tune: “Vinyl Memories” by Mikael Manvelyan.
Assorted sound effects: AudioBlocks.


Great podcast!
Please thank Carole for the heads-up for ”How to Win Friends and Disappear People”. I love these kinds of podcasts altho when I was a youngster (in a previous millenium) we refered to them as "radio plays".