Podcast
Dark Web Intelligence and the Access Broker Economy
Cybercriminals rarely go straight from initial access to a ransomware attack – there’s an entire ecosystem that makes it possible, and a window to stop it. Joan Woodward, President of the Travelers Institute, was joined by Josh Doguet, Senior Manager of Threat Intelligence at Travelers, to explore how dark web intelligence can disrupt the access broker economy and stop ransomware before it starts.
Josh explains how criminals gain initial access and then sell it to the highest bidder and how organizations can cut them off before they reach the lucrative ransomware stage. His conversation offers a behind-the-scenes look at proactive threat intelligence and the practical steps every organization can take to get ahead of the threat.
This episode is part of a special cyber-focused series of the Travelers Institute Risk & Resilience podcast. This program is presented as part of the Travelers Institute’s Cyber: Prepare, Prevent, Mitigate, Restore® initiative, which promotes dialogue and education to help leaders prepare for and respond to cyber incidents.
JOAN WOODWARD: Hi, folks, and welcome back to the Travelers Institute Risk and Resilience podcast. I’m your host, Joan Woodward, president of the Travelers Institute. Cyber threats are evolving fast, and staying ahead of them has never been more critical for organizations of every size. This special four-part cybersecurity series brings together cyber experts from across Travelers to unpack the issues that matter most for organizations navigating today’s threat landscape. It’s presented as part of the Travelers Institute Cyber: Prepare, Prevent, Mitigate, Restore initiative, which promotes dialogue and education to help leaders prepare for and respond to cyber incidents.
JOAN WOODWARD: In this episode, I was joined by Josh Doguet, Senior Manager of Threat Intelligence at Travelers, to really explore how the underground marketplace for stolen credentials works and what Travelers does to get in front of it. Take a listen.
Josh, so happy that you’re here today, because we’re going to be talking about the dark web, something that is definitely a little scary and mysterious to some of our listeners today. Feels like it fits this time of year with Halloween approaching. Let’s start by explaining what the dark web actually is. And how did you end up working in this corner of cybersecurity?
JOSH DOGUET: Sure. And thanks for having me on, Joan. I am a lawyer by trade, but before law school, I’d had a few different technical roles in network operations and web development at the IT help desk. Nothing fancy, but I spent enough time in those roles to understand how networks and applications fit together. And that was enough to make me want to eventually move back in a technical direction. So when an opportunity came along to work on the cyber risk prevention and mitigation side of things, I didn’t think I could pass that up. My job now sits at the intersection of threat intelligence and claims analytics. So, on one side, I am tracking the threats that our policyholders face, and on the other, I’m analyzing what actually happens when those threats materialize. And this work not only informs how our team advises our policyholders and what risks we’re alerting them to, to try to prevent those incidents from happening in the first place, but it also informs several other parts of the business as well, and I find that very rewarding. As to the dark web, the first thing to understand is that the dark web isn’t something that you can visit from a regular web browser, and that is because all of the websites that together make up the dark web aren’t located on the ordinary internet. They’re located inside “dark nets,” which are also known as anonymity networks. And these are overlay networks. They run on top of the ordinary internet. And as the name implies, their entire purpose is to anonymize the activity that
takes place on them, to hide the websites they contain, to hide who is operating those websites, and to hide who is visiting them. There are several of these anonymity networks, but the most popular one, by a wide margin, is Tor, which stands for The Onion Router. And when someone says the dark web, they’re almost always talking about websites hosted on the Tor network, whether they know that or not. And to reach those websites, you first have to connect to the network, and you need specific software to be able to do that. Because, like I said, if you type a dark web address into a regular web browser, you are not going to get anywhere.
JOAN WOODWARD: Wow. Okay. So for most people listening, that’s really a lot of new information, at least to me. I want to break it down just a little bit more. So is the software one of the keys? Is that what makes it anonymous, then?
JOSH DOGUET: Yeah, exactly. So it’s a combination of the software on your computer and the software that runs on all of the nodes that route your traffic around the network. Normally, when you visit a website, you are making a direct connection, so your internet provider can see where you went, the website can see where you came from. And those two facts taken together can basically be used to identify you. Darknets – like Tor – they break that connection between the user and the destination. So when you use the Tor browser, it wraps your traffic in multiple layers of encryption, hence the onion metaphor. That encrypted traffic then gets routed through several nodes on the network before it reaches its ultimate destination. And because of how those layers of encryption work, each node on the network only ever sees one piece of the picture. So the first node that your traffic gets routed through can see who you are. But it can’t tell what site you’re visiting. The last node can see what site you’re visiting, but it can’t see who you are. So no single point in the chain ever knows both. That is glossing over a lot of detail because I don’t want to get too far into the weeds, but that is onion routing in a nutshell.
JOAN WOODWARD: Okay, so let’s go back to the onion router. If I leave our conversation now and download Tor and just start browsing, am I browsing the dark web? And for the record, folks, I’m not going to do any of that. I’m just curious.
JOSH DOGUET: So you’re not necessarily browsing the dark web. And that’s a useful thing to clear up. You can still visit ordinary websites while using Tor. And because of the way the traffic is encrypted and routed, your browsing will be anonymous. And in fact, most Tor traffic is just people using the regular web anonymously. That could be journalists trying to protect sources. It could be people in countries with heavy government surveillance. Or it could just be privacy-conscious people who don’t want to be tracked. Only a fraction of Tor users ever visit websites that are hosted entirely within the Tor network. And it’s those sites that actually make up what we call the dark web. These are sites with those really long, weird-looking addresses that end in .onion – the kind of URL that you might find in a ransom note. When you visit a regular website using Tor, you’re getting anonymity, but your traffic is still ultimately exiting the network to reach its destination. When you’re visiting a Tor-hosted site, your traffic never has to exit. It stays inside the network from end to end. And this means that both ends of the connection are hidden, not just the visitor, but the server, too. Nobody can tell where that machine is physically sitting. And that is exactly why the dark web is one of the main places you’ll find cybercriminal marketplaces. The operators of those sites want their infrastructure to be anonymous, just like their customers want to be anonymous.
JOAN WOODWARD: Okay. So what I’m hearing is just because someone is using Tor, it doesn’t mean they’re up to no good. And I’m assuming that those who are up to no good, like cybercriminals, are not only on Tor or the dark web. So is that right?
JOSH DOGUET: Yeah, that’s exactly right. A lot of what most people picture happening on the dark web isn’t happening there anymore. Or at least it’s not happening exclusively there. It’s also happening on messaging apps like Telegram, which some listeners may be familiar with because it’s a fairly mainstream and widely used messaging app. And while it does offer encrypted messaging, in a lot of ways, it’s the opposite of the dark web being that it’s run by a single company on corporate servers. And Telegram accounts are tied to phone numbers. It’s not an app that was ever built for anonymity in the same way that Tor was.
JOAN WOODWARD: Okay. So then why would a criminal use one of those messaging platforms then?
JOSH DOGUET: For convenience, I’d say mostly. Telegram gives them broadcast channels that let them reach lots of subscribers instantly. It gives them large private groups. It gives them these bots that can handle sales and delivery in an automated fashion. Standing up a website on Tor takes effort, while opening a Telegram channel doesn’t. So for a lot of this activity, Telegram is simply more efficient. Criminals are giving up some amount of anonymity to get that convenience. But for a high-volume business, that is a trade that plenty of them are willing to make.
JOAN WOODWARD: Okay. Got it. So then it’s a business decision, right? And they’re thinking about the ROI or efficiency. So whether it’s on the dark web or something like Telegram, what are cybercriminals? I mean, what are they actually doing there?
JOSH DOGUET: So I would probably split the overall activity into two buckets. In the first bucket, I would put the things that this audience is probably already most familiar with. And that is, the dark web is where ransomware groups host their leak sites and their negotiation portals. So if you are the victim of a ransomware attack, you’ll likely receive a note containing an address to a site on the dark web. If you log in to that, you’ll basically get a customer service chat window, which you can use to communicate with the attackers. There, they will try to prove that they’ve stolen your data and that they can decrypt your files. And if you decide to pay, that’s where you’ll haggle over the price. And if you don’t pay, then the group is going to disclose your name on their leak site and eventually publish the data that they’ve stolen. Now, while these sites are on the dark web, they’re still technically public. So they’re a major source of intelligence. For everyone who works in this space, we can monitor them and use the number of postings, as sort of like a proxy for overall ransomware activity, to assess things like who is being targeted, at what volume, and how the ecosystem is generally changing over time.
JOAN WOODWARD: One of the things I’ve learned in doing our cyber series that really surprised me is that these criminals have set up the logistics and the platforms, frankly, of a legitimate business like a customer service window. I mean, they’re running sophisticated operations here. Okay, you said two buckets, so don’t hold back on us. What’s the second bucket, then?
JOSH DOGUET: Right. So in the second bucket I would put what I’d call the marketplace and trade show type of activities. This is really the heart of the economy, and it takes place in forums and group chats and broadcast channels across both the dark web and Telegram. I call it a trade show because that’s what it resembles. This is where threat actors share tradecraft. It’s where they recruit others to join their groups and coordinate their activities. And it’s where they advertise and sell things like tools, malware exploits, stolen data, stolen credentials. It really runs the gamut.
JOAN WOODWARD: Wow. Trade shows. So another one of those terms that one normally hears in the course of a normal business channel. So they’re selling these tools and data to threat actors, then.
JOSH DOGUET: Exactly. It’s important to understand that there’s an entire division of labor that is underpinning this ecosystem. So gone are the days where, if you wanted to conduct a cyberattack, you had to be good at everything, whether that’s writing the malware, infecting machines, breaking into the environment and so forth. Now, every almost piece of that chain is something that you can specialize in and then rent to someone else or rent from someone else. Starting with probably the best-known example, there’s ransomware as a service. So there’s a core group of developers who write and maintain the ransomware and operate the associated leak site and the negotiation infrastructure. And then there are these less-skilled affiliates of those operators who access the victim’s environment and deploy the ransomware. And there’s usually some form of revenue sharing between the two: commonly something like the affiliate keeping 80% of the ransom and the operator taking a 20% cut. There’s also phishing as a service. So developers will build entire platforms for managing and automating phishing campaigns. These platforms come with things like ready-made fake login pages, email templates, dashboards and functionality built specifically to get around multifactor authentication. It’s like an entire SaaS product – a software as a service product. And attackers will pay a subscription fee to use the platform. And they’ll have everything they need to carry out a business email compromise, usually with the goal of rerouting a wire payment to their own bank account. The last of these that I think is worth mentioning is credential-stealing malware, which is often called “info stealers” or just “stealers.” So these are fast, lightweight pieces of malware that are designed to infect a machine, harvest everything of value on it.
So that’s things like safe passwords, session tokens, browser data, and then they send all of that back to the attacker. And it’s the same case here. So there’s someone who is developing the stealer malware. Then there is someone else who is renting access to the malware and then deploying it and then selling the credentials that it collects, which usually come in the form of credential logs from each infected machine. And then there’s yet someone else who buys those credentials with the intention of either using them for an attack or with the intention of reselling them to someone who does.
JOAN WOODWARD: So that really sounds like a middleman to me. In a typical business arrangement in the cyber world, that’s the access broker, then, right?
JOSH DOGUET: Right. That’s right. Access brokers frequently purchase logs, like this, of stolen credentials. And they’re looking to find and resell the highest-value credentials from those logs.
JOAN WOODWARD: Okay. So that’s a good segue, then. So let’s talk about access brokers. Can you give us some details on their business model and what they’re selling and who are their customers?
JOSH DOGUET: Sure. So as the name implies, access brokers are selling access. Namely, they’re selling remote access credentials to corporate networks, which would include things like employee or administrator logins for VPNs, for remote desktop, web credentials, single sign-on credentials, cloud console credentials – basically anything that can be used to gain an initial foothold in a local environment or a cloud environment. Their whole business model revolves around acquiring these credentials and then selling them to other threat actors, usually for a few hundred or a few thousand dollars each. At the end of the day, the pricing depends heavily on things like the victim’s revenue, their industry, their country, along with the level of access that is involved. So in other words, admin credentials for a large U.S.- based manufacturer that can’t afford any downtime would likely fetch a much higher price than, let’s say, a user login for a small retailer. Access brokers are selling these credentials primarily to ransomware affiliates, and I think it’s worth pausing for a second on why that arrangement suits both sides. So rather than taking on the effort of conducting the attack, along with the risk of an uncertain payout, the access broker is running more of a steadier, volume-based business. Meanwhile, this allows the ransomware affiliate who is purchasing the stolen credential to breeze through what is often the hardest and most time-consuming part of the attack, which is breaking into the environment. So it all goes back to what I was saying a moment ago about specialization within this ecosystem.
JOAN WOODWARD: So how do these access brokers then get their hands on those credentials in the first place?
JOSH DOGUET: Yeah. So a minute ago I mentioned info stealers – that’s the malware that infiltrates safe credentials from infected computers. That is one major source for them. So an access broker may have a subscription to access info stealer logs in very large batches – what’s generally called a cloud of logs. So brokers are going to sift through those logs looking for the needles in the haystack, because there’s a lot of noise in those logs – someone’s streaming service or social media login that is not particularly valuable, and there’s going to be plenty of that type of thing in these logs. But a login for the admin console of a VPN appliance, that is the needle. The other major avenue for access brokers is going after targets directly. That could mean phishing employees for their credentials. Or it could mean going after internet-exposed remote-access appliances like VPNs and firewalls. So with respect to the latter, they might exploit an unpatched vulnerability, but that takes time and that takes effort, so more often they’ll just engage in old-fashioned password guessing – what we call brute forcing – which can be heavily automated these days and is far more effective than you might think, especially when the victim isn’t using multifactor authentication. But regardless of whether these credentials are coming from brute forcing or whether they’re coming from exploitation or from theft, the next step is really the same. Once an access broker has a corporate credential, they’re going to validate it. They’re going to log in. They’re going to confirm that it works. They’ll try to figure out what level of access they have. And then they’re going to package everything up and they’re going to list it for sale.
JOAN WOODWARD: So now I want to bring this back to the people and organizations that are listening in today. A stolen password almost sounds mundane, really, next to some sophisticated software exploit. Why is this more dangerous than it sounds?
JOSH DOGUET: It’s dangerous precisely because it’s mundane. What makes the use of stolen credentials by ransomware groups so effective is that it doesn’t give off any of the signals that you are under attack, so no malware is being executed that could trigger an alert. Nothing is being exploited. The attacker is just logging in with a valid username and password, so they just look like an ordinary employee who is going about their business, and by the time anything looks wrong, they’re already inside the environment. And I would say it’s also dangerous because of its prevalence. So the exact numbers will differ a bit from report to report and year to year, but I think it’s pretty safe to say that stolen credentials are one of the top initial access vectors across incident types, and that is why monitoring this ecosystem for the sale of our policyholders’ credentials matters a great deal. There is this gap between the moment a credential is listed for sale on the dark web or on Telegram and the moment that an attacker actually uses it. So if you know about the listing, then during that window, you can do something about it before the attack ever happens.
JOAN WOODWARD: So can you give us an idea of how that monitoring actually works? How do you catch an access broker who is attempting, then, to monetize a policyholder’s credential?
JOSH DOGUET: So our team, the Cyber Risk Services team, we partner with a dark web intelligence vendor that specializes in exactly this part of the ecosystem, the access broker market. And what makes them valuable isn’t some kind of dark web scanning platform; it’s their people. They have, essentially, undercover operatives who maintain personas in these communities, and those personas present themselves as ransomware affiliates who are in the market for access, and they are not using accounts that they just spun up last week. Building a credible identity in this world takes time because these communities run on reputation. You have to be vouched for, essentially. You have to have a history. You have to talk like someone who belongs there. And so these operatives have invested a lot of effort in establishing that standing, and that standing changes what they can see. They are monitoring broker listings as they get posted, for sure. But because they have these established reputations, access brokers also come to them directly. A broker with a new offering will sometimes bring it directly to buyers that he trusts before it ever gets listed for sale somewhere more public, which means that we are sometimes identifying the threat at the earliest possible moment. So whenever our partner determines that one of those listings is for a policyholder, they pass all that intel along to us and it becomes the job of our team to let the policyholder know. So we’ll issue an alert to them via email, but we also work really hard to get somebody on the phone, because we definitely want to make sure that we have their attention. And the first thing that we’re doing is validating the threat. We want to connect what is in the listing back to something real in their environment. So is this an account that exists? Is this a system that is actually theirs? Because we don’t want to send a company into an emergency response over a false alarm, right? But we also don’t want them to dismiss something real as just noise. So validation is the first step, and if that checks out, then we walk them through what they need to do about it. That would involve things like rotating the credentials, revoking the active sessions, finding out how the credential was compromised in the first place. For instance, if there’s malware on a user’s machine, and if there’s any indication of lateral movement beyond the initial access. In most cases, we’re really going to encourage them to file a cyber claim.
JOAN WOODWARD: Really. To file a cyber claim, even though nothing’s happened yet.
JOSH DOGUET: Yeah, that’s the thing. So if the access listing is valid, then something has happened, right? When we say that we’ve caught this early, what we mean is that we’ve caught it before ransomware was deployed, but remember that brokers typically validate the access before they list it. And that means someone logged in, confirmed the credential worked and very possibly looked around to see what was reachable. So there may already have been activity within the environment. The process of investigating whether that kind of lateral movement has occurred and also trying to figure out how the credential was compromised in the first place – that may require expertise that the policyholder’s internal team doesn’t have. Moreover, I would say this is a criminal ecosystem, and you can’t definitively rule out that the broker hasn’t already sold the credential to another ransomware affiliate who is already in the process of using it even if that broker claimed that the sale would be exclusive. So opening a claim brings in a digital forensics firm that can do all of this heavy lifting. So the calculus for the policyholder, I’d say, is pretty favorable, and that would be a bounded investigation now versus the possibility of a full ransomware event later on.
JOAN WOODWARD: So I guess that makes sense, then, for the organizations listening to this. What should they do with this information?
JOSH DOGUET: Well, I would start by reiterating that dark web intelligence and alerting is a benefit that we currently provide to our cyber policyholders, so that is one way of accessing this type of service. If that’s not an option, then whether this kind of service fits within the budget is really a decision that every organization has to make for itself. What I would say is if you’re going to do it, do your due diligence and know what you’re actually paying for. There are a lot of vendors out there offering something that they call dark web intelligence or dark web monitoring, and the quality of that service offering varies enormously. So there are a few things that I probably ask about. First, where does the data actually come from? There is a big difference between a provider that is pulling fresh information out of an active criminal marketplace and one that is primarily repackaging old breach databases. Relatedly, ask what they do about noise. So old breach data gets recycled and resold constantly. So a low-quality service will happily alert you to a password that your employee reset years ago. If your team gets buried in alerts like that, they’re going to start ignoring all of them, including the ones that actually matter. So it’s important to ask whether a human validates findings before an alert reaches you. And finally, I would say if a vendor tells you that they can remove your data from the dark web, I would be very skeptical of that claim because that’s largely technically not possible.
JOAN WOODWARD: So then for organizations that aren’t going down that road, what else can they do?
JOSH DOGUET: They can do several things, I’d say. So we talked about brute forcing earlier – that’s attackers using automated tools to guess their way into remote access. So MFA (multifactor authentication) is the first line of defense against this since it adds an additional layer of protection to the username-password combo. Ideally, phishing-resistant MFA should be used, since it relies on cryptographic protocols rather than codes that attackers can intercept. If you can’t roll out MFA everywhere at once, I would prioritize the two places that matter most, and that would be anything internet facing, like remote access systems, and also your administrator accounts, because access to those systems is exactly what brokers are advertising. We also talked about attackers exploiting vulnerabilities in internet-facing devices. So you need to run an effective vulnerability management program, which is to say that you need to genuinely act fast on critical vulnerabilities in those edge devices – things like VPN appliances, firewalls, anything sitting on the perimeter. Really, the window between a vulnerability being disclosed and it being exploited is now very short, and those devices are the ones the attackers are really looking at first. We talked about credential-stealing malware, and I would say that the answer there is a modern EDR solution – that’s endpoint detection and response. You want an EDR solution with behavioral analysis capabilities and with somebody who is actually watching it around the clock. Those info stealers are fast and they’re quiet. If your solution can’t detect them, or if it generates an alert at two o’clock in the morning that no one is going to read, then that solution really isn’t protecting you. And I’d also say that type of coverage is what catches the anomalous activity that follows, so someone logging in from a place that your employees don’t normally work at a time they aren’t normally working. Finally, I would say you need to have a plan for when credential theft happens anyway – whether it’s from an info stealer or from a phishing email or otherwise – because it will happen. Resetting the password is not enough. You need to also revoke any active session tokens because those get stolen too. And a stolen session token can let someone right back in even after a password is changed. And like I was saying earlier, you also need to find and remove the malware that caused it. If you miss any of those things, then you haven’t actually closed the door as a whole. These recommendations might sound a lot less alluring than dark web intelligence, but you have to remember, like we talked about, these threat actors in this ecosystem, they are running a volume business and they are looking for easy wins. Every one of those controls that I just talked about, they make you a harder target, so they matter more than you might think.
JOAN WOODWARD: Well, Josh, I think that’s just a great place for us to stop. I always like wrapping up with some practical steps anyone can take, but also really appreciate your guided tour of the dark web and, really, help. Importantly, you and your teams are being proactive within this space. It’s really remarkable. So a big thanks to everyone listening in. I hope you can catch the other episodes in our miniseries on cybersecurity, including our soon-to-be-released finale, where we’ll share even more cybersecurity guidance. So Josh, thanks again and thanks to all my listeners out there.
JOSH DOGUET: Thank you for having me.
JOAN WOODWARD: Thank you for joining us for this insightful conversation. Check out our show notes for more information about the Travelers Institute and to sign up for our mailing list and LinkedIn newsletter. If you enjoy our content, please be sure to subscribe, rate and review on Apple Podcasts and Spotify. You can also contact me by following me on LinkedIn or emailing [email protected] to let me know what you really think and what else you’d like to see from us and hear from us at the Travelers Institute. Thanks again so much for listening.
This episode is also available on Apple Podcasts®.

Host
Joan Woodward
President, Travelers Institute; Executive Vice President, Public Policy, Travelers
Key takeaways
- The cybercriminal ecosystem has evolved into a sophisticated, highly specialized marketplace where threat actors can buy, sell and rent everything from stolen credentials to ready-made attack toolkits – much of it operating on the dark web and on mainstream messaging apps.
- Access brokers – criminals who specialize in acquiring and reselling corporate login credentials – are a key link in the ransomware attack chain. Stolen credentials are consistently among the most common ways attackers get into an organization’s environment.
- A stolen credential is more dangerous than it sounds precisely because it doesn’t look like an attack. Someone logging in with a valid username and password generates none of the signals that come with malware or exploitation.
- There is a window between the moment a credential is listed for sale and the moment it gets used to launch an attack – and organizations that know about a listing during that window can act before an attack happens.
- Strong multifactor authentication, fast patching of internet-facing devices and a modern endpoint detection and response solution aren’t flashy controls, but they directly address the methods access brokers rely on and make organizations significantly harder to compromise.
Travelers Institute Risk & Resilience podcast
Subscribe on LinkedIn® (Opens in a new window)
The Travelers Institute Risk & Resilience podcast is available on Spotify and Apple Podcasts®. Subscribe today to be alerted when new episodes are released.
Related content
Episode one
AI’s Impact on Cyber Vulnerability and Defense
Lauren Winchester, Head of Cyber Risk Services, Travelers, joined Jesse Sugarman, AVP, Travelers Institute®, to explore how AI is reshaping vulnerability discovery, detection and patching – for both defenders and attackers – and what steps security teams can take to stay a step ahead.
Episode two
Closing the AI Governance Gap
Lauren Winchester, Head of Cyber Risk Services, Travelers, joined Jesse Sugarman, AVP, Travelers Institute, to explain what the gap between AI adoption and governance maturity means for businesses – and how leaders can get ahead of the risk.
Episode four
A Cybersecurity Wish List for Businesses of Any Size
Joan Woodward, President, Travelers Institute, sat down with John Menefee, Vice President and Enterprise Cyber Lead, Travelers, and Fayon Atkinson, Senior Manager, Cyber Risk Services, Travelers, to explore recommended actions and guidance for businesses and why they matter in an era of evolving AI-driven cyber threats.
In the news
PropertyCasualty360
Digital Insurance
UserFriendly 2.0
Join our mailing list
Invitations, event announcements, replays and more — delivered to you.
Sign up
