Overview
Patching vulnerabilities by CVSS and tackling tickets on a first-come, first-served basis results in high-priority issues being left unresolved until it’s too late.We’ll follow a fictional company through 24 hours and show where disconnected tools cause real misses: a public-facing asset that appears just after the daily scan, and a mid-severity vulnerability whose true risk only becomes clear when you see it advertised for sale by an initial access broker.
We’ll show how your threat intelligence and attack surface management programs can combine to minimize blind spots and misprioritization by seeing both sides: your external exposure and what attackers are actually doing.
Key Takeaways
- How exposure shifts across time zones, and what scans miss between runs
- How disconnected threat intelligence and ASM lead to blind spots
- How to re-prioritize based on what attackers are targeting, not just severity scores
Full Transcript
Alex Blackman: Hello everyone and thank you for joining us on today's webinar where we're going to learn why tackling exposures in the order they reach your queue can leave the riskiest ones buried and unresolved. So to show you how to solve this across both the threat intelligence side and ticket queues – the ASM side – we have two expert speakers with us today. First up, representing the CTI side, we have former US Army security specialist Andrew Craig. And representing exposure management, we have Rachel Schwalk, former analyst, forensic examiner, and detection engineer.
So, how can teams actually prioritize what matters and what's real? To show you this and to bring it to life, we are going to follow a fictional company through 24 hours to see what happens when you have disconnected tooling and processes. And importantly, we're going to look at how you become more preemptive – you can fix those issues – and less reactive.
We've also packaged a load of great resources in the attachments area as well, so please do check those out. Everything from a personalized dark web risk report to our latest and greatest preemptive handbook, preemptive buying guide, checklist of questions, everything you need to know when making this shift from reactive to preemptive.
But with that, I'm going to hand us over to Rachel for morning and the start of this story on the ASM side of our fictional company. So Rachel, it's across to you.
Rachel Schwalk: Yeah, thanks Alex. Yeah. So I am playing the role of a SOC analyst, right? So I am a SOC analyst for a global organization. I'm dealing with SIEM alerts, tickets... you know, these don't ever stop, right? I'm also handling things like user-reported phishing emails. I'm also triaging malware, right? Like somebody clicked a bad download. And on top of all of that, managing the exposure queue. In the tool set, we do have an internal vulnerability management solution and an external vulnerability management solution. We have internal scans that run weekly and an external scan that runs every 72 hours. And typically, things that aren't like absolutely critical, they're going to take 30 days to resolve, optimistically, right? It has to go through that change management process and company processes to get those fixed.
Blind spots do happen, especially in a global organization. We have developers that are across the world. They're spinning up staging servers; they're doing things they need to test like database migrations, software upgrades, whatever the case may be.
So in this scenario, one day a developer does that exact thing, right? They spin up a staging server. It is running SSH. It doesn't go through our normal approved processes. It's supposed to be temporary, just a test server. So, I don't know about it; the rest of the security team doesn't know about it. That box doesn't get any monitoring going to the SIEM; there's no EDR agent on it. There's whatever we have in CloudTrail basically. So if we did go and investigate in CloudTrail, we could see some things. But it also didn't have security group segmentation from a production instance. So not ideal.
But in this scenario, I don't know about this specific box yet. But I do see from our external attack surface management that we might be vulnerable to a moderate CVE – like a 6.5 CVSS score basically – and it could allow MFA bypass. You have to be authenticated, right? And so this could be any vulnerability, though, that is medium that you need to be authenticated. So I'm immediately in my mind deprioritizing it because of needing to be authenticated, right? Dealing with other incidents, it's not a critical. I don't know about this test server yet that I might would have been more suspicious of; it could have had default credentials or weak credentials.
And so, also because I'm not seeing it yet, our external scans are every 72 hours. So basing this CVE off of something that I do know about being vulnerable, I am going to pass it to Andrew so he can talk more about what he sees on the intel threat analyst side of the house, and he can go into more about what he sees on initial access broker postings, that kind of thing.
Andrew Craig: Right, so moving forward from the cyber threat intelligence side of things, right? As you see in the slide, that is an actual initial access broker solicitation of a type of access, right? A lot of times those initial access brokers will act as middlemen before a ransomware and/or a breach of some type, right? But nevertheless, the threat intel team, right, while Rachel's team is working on internal tickets, the CTI is looking outward tracking ransomware groups and initial access brokers, any dark web chatter, Telegram chatter, stuff like that.
So by midday, right – and this can be even more severe depending on time zones – after sifting through all these industry reports, going through reporting, going into these environments like XSS or Exploit or Telegram, any of these things, having a way to track and monitor, you can see that SSH is being weaponized currently. So if an SSH exposure is currently being targeted based off these threat groups, that should increase the prioritization, right? We're sitting at a 6.5 roughly, right?
So in a healthy top-down CTI workflow, my job is to push that intelligence down to the SOC so they can prioritize what matters. We drop that intel ticket into a queue, right, warning that SSH is currently being exploited by said groups. From there, you can do threat intel associated to the groups if there are any or if they're active, right? How are they pivoting? What is the exploitation? What tools are they using? So on and so forth. IOCs – anything that could be correlated to that.
Now with that, the intelligence and the exposure live in two completely different platforms typically, right? So if you're doing deep investigations, that's going to be an entirely different platform; it's going to be a manual process, right? So, my team knows attackers are hunting for the exposed SSH at that point, and Rachel's team has a 6.5 rating on the CVE, right? In their backlog, the 6.5 is not a priority. But neither team knows they're looking at the exact same thing and that they could possibly be vulnerable, because that connection has not been made yet.
Right. So, moving forward... next slide, please. Moving forward, the CTI team will continue that investigation, right? So say there was a breach of credentials that day. It could be historic – historic breaches are just as important as new breaches. A lot of times new breaches are just regurgitations of old breaches, but new ones do happen; they do happen quite frequently. Nevertheless, once we start monitoring – and this can be via monitoring capabilities or if you're doing it in a manual way – you see that admin credentials are being exposed to different SSH environments in which you are running, or an individual within the team is reusing passwords and that password has been associated to that. Now that criticality score is going to increase, right?
Moreover, an attacker doesn't need a zero-day to access said network. If those account credentials are being reused and exposed in a manner in which a lot of us have, then that vulnerability is going to increase in exploitability.
Now, looking at that ticket queue, going back, the CVSS score hasn't changed, right? It's still at a 6.5 based off how we're monitoring. What has changed is the real-world likelihood of exploitation.
So we have now three critical signals: an unapproved server, the active attack interest by threat actors, and exposed credentials that could lead to that initial access being granted sitting in three separate silos, generally speaking. Because if nothing is being dynamically reranked, that ticket isn't going to move up in prioritization. And in theory, if that is true... I mean, if we learned anything from this past week with everything that's going on, it's how quick the environment can change, right? So without having this dynamic approach to ticketing, what could look like a very low risk in reality could escalate quite quickly if you don't have the entirety of that threat picture at your disposal. Next slide.
Alex Blackman: Well, Andrew, just quickly before we move on from that one as well, I was interested to get your thoughts on this because I've certainly heard mentioned in the past that CTI can be deprioritized in organizations, and maybe some of the fine... you know, you've just presented an active, real-life IAB listing that doesn't always get the attention it should do. Do you have any thoughts on why CTI isn't always getting the love and attention that it can bring to an organization?
Andrew Craig: Yeah, I think the CTI thing... not that it's newer, but it's becoming more mainstream, right? So going back to this, we're taking more proactive approaches in cyber defense, right? In cybersecurity, right? And that's our model; that's what we want to do, and CTI is a big part of that. Up until recent years – and this is going to change, it's a dynamic world – as more attacks are happening, as that mean exposure needs to be cut down, how quick people can exploit vulnerabilities in which they're finding with AI and whatever else, CTI has been low on that priority list because they've got to justify the threat and the overhead they're going to pay for a team, not to mention the training and stuff like that.
So a great stopgap is absolutely having a dark web monitoring capability or something of the sort, right? And you can pipe those into SIEMs, into SOARs, and whatever that may be. But a big thing is, one, it depends on the size of the organization. If you have very mature security stacks, it's a lot easier to start implementing CTI teams. If you fall within that middle to lower range, how do you justify hiring a team in that manner? Not to mention, if you contract out your SOC and all these other factors, now where do you sit in that organization? How do you prioritize? When are you truly at threat?
That's the mentality; the complacency within the industry is, "Am I truly at threat?" Like, I'm not a Booz, right? I'm not a Delta Air Lines; I'm not United. It doesn't really matter because a lot of these guys are going after the lowest-hanging fruit. So if there is exposure... and this is even more critically important to environments in which it is mandated by government that you protect said data, right? Like client data, financial records, any of that, that becomes critically more important because you need to have a more proactive approach. And in lieu of having a full CTI team, at a minimum, you would want some type of means to do automated collection to start piping into the ticketing systems and the alerting systems accordingly.
Alex Blackman: Brilliant. Thanks, Andrew. I'll jump us ahead to the next slide.
Rachel Schwalk: All right. So, this is moving us back within the same 24 hours, but if we had those connections together, right? So say we have that external scan occurring every hour. See that box that becomes live; we've got threat intel matching that asset. So those technologies that Andrew was showing, we were like, okay, that could be us, right? And then also the fact of those leaked credentials that Andrew saw out as a listing too... that makes it to where those three separate items become one when we're able to link it together.
Andrew Craig: I think another great way to envision this – and this is coming from the CTI side, right? Rachel and I have very different backgrounds, but it's great that we can mesh those together, especially when we work currently, right? The way I kind of look at this is a top-down enrichment approach, right? So in that situation, as soon as an SSH asset is discovered, the CTI will automatically start mapping the external threat that is correlated to that asset, right? Instead of like a generic vulnerability alert that the SOC sees, right?
Secondly, you have a dynamic reranking system kind of like we just spoke about a second ago, right? So, say you get there in the mid-afternoon and Rachel's off work, but I'm working because I'm on the West Coast. Well, when those leaked admin credentials surface on the dark web or you find them because you continue your investigation, they cross-reference against identity providers and map against the assets which we are then being alerted to, right? That should inherently increase the severity and risk score associated to the vulnerability.
And then lastly, that bottom-up feedback. So when a SOC analyst immediately isolates and remedies a server, well within 24 hours, that should come to the CTI team. So we can then start going out and seeing what else in the infrastructure matches exactly what they just shut down. What if there's other assets?
And moreover, before the whole process restarts, going through that intelligence workflow and that intelligence cycle, we go back through that cycle and figure out where we are, who's actively targeting it, and what on our periphery is actually meeting the parameters in which threat actors are hunting for.
Alex Blackman: Thanks Andrew. And yeah, you make a really good point about time zones. Assuming everyone is on the same... we're not; people take vacation. If the systems are siloed, something really critical could be buried, if it's even scanned at all. Rachel said a 72-hour scan; to me, that feels glacial given the speed of the attacks that we've seen lately. So not only do you need regular scanning, but you need that interconnectivity between teams and systems, and the ability to combine rankings really if you do want to get ahead of these threats.
Andrew Craig: Right. I think that's like the big difference, right, between drowning in alerts and preemptive security. That's really what we want to focus on. So when you intersect – confirm exploitability from your exposure data and observe attacker interest from your CTI capability or your CTI team or your monitoring, whatever that may be – your SOC stops guessing. So you can both work together and fix and triage the exposure in a more preemptive front to try to prevent the possibility of breach. Right.
Alex Blackman: Yeah, and that's exactly what we're looking at here. You know, it's a very simplistic diagram that addresses a very complex approach to cybersecurity. Seeing both sides is how we'll refer to it at Searchlight Cyber. It's no good having two teams doing intelligent work in silos; you do need to combine those to prioritize what's right at the top of your queue. You certainly should not be working through the queue on a first-come, first-served basis because it could bury those most critical issues. As we discussed at the beginning, the SSH server that Rachel identified would have been the most critical of the day. Were it first-come, first-served, it would not have been actioned given the 6.5 severity rating.
And so with that, we are pretty much at the end of the session. I just want to highlight as well that although our session today is Searchlight Cyber's view on the world, we are certainly not alone in thinking this way. So Gartner predicts that in 2030 – which is only three years away, it's really not that far – the spend on preemptive threat exposure management (PTEM) is going to increase from just 5% in 2024 to 50% of all security spend. So we are very much seeing a shift towards this preemptive approach. What we've just detailed is how we believe that you can address this.
But with that preemptive call to arms, if you will, I'm going to open up the floor to questions. So please feel free to use the question mark symbol on the right-hand side. Use the attachments as well to download resources that support everything we've covered today, and I'll start going through the questions.
So first up we have a question here. This one is on combining the sources of threat intelligence. So the question is: "How do your customers integrate threat intelligence, like dark web monitoring for initial access brokers that we showed, into the ticketing queues without overwhelming SOC teams?"
Andrew Craig: Right, so I think there's multiple ways, right? Because if you have CTI, you're always going to want some type of monitoring capability, right? And most of those monitoring capabilities – we're no different with our offerings – are going to have the ability to integrate into your SIEM scores. That's only part of the puzzle, right? So with a lot of the monitoring capabilities, you'll absolutely be alerted, but then you're not going to be able to pivot to get that finite detail associated to the threat.
Say if your credentials are republished on a combo list within a Telegram channel, you would absolutely be notified typically within the hour of, "Hey, this has been posted." The issue with that is: how do I know that it's current? How do I know this isn't a regurgitation? Did I already fix it? So then you're going to have to develop the workflows, and again, I think we do a very good job of being able to integrate the technologies to start automatically triaging a lot of those credentials.
Moreover, say in the event that you suspect there's an initial access broker – a very niche example here – in the event that does happen, how do I start looking at the individual soliciting said access? Or say you're mentioned on a Pastebin with all your sensitive server information... happens all the time. How do I pivot on the individual in question? Is it a credible threat? Is what they posted true? So on so forth. If you're not in all those environments, you can't really see it.
So there's always going to be a means. I think this is true with anything, especially as we're getting to the AI era: having that human in the loop to verify. It becomes very easy and we become very complacent when it comes to everything AI. However, having someone there to verify some of the information, dig a little bit deeper, is always going to be worth the extra time.
Alex Blackman: Thanks, Andrew. Next question is probably for you, Rachel. This one is on ASM. Given that Andrew is on the CTI side, this is from somebody whose current ASM scans daily. They believe they already have more alerts than they can get through. How do teams manage when that scan is hourly? So I guess it's very similar to the sort of question we just had.
Rachel Schwalk: Yeah, I think with the hourly scans though, you're not overwhelming teams as much as you're alerting when items that wouldn't be caught during that typical 24-hour scan would come up. A lot of times, we were talking about the different time zones. What if boxes are only coming up, you know, 2:00 a.m. my time, and that scan is set to go at 8:00 a.m. my time and feed my tickets in? You would never know about that risk that happens overnight.
And especially if you have that validation of exploitability. It's not just version checking of like, "Oh, this could be something that you are vulnerable to," but having that confirmation of "This was tested and confirmed exploitable." It's really imperative that you have that hourly look into things because otherwise there could be major blind spots that occur.
Alex Blackman: Yeah, we spoke about ephemeral assets when we were planning this webinar, didn't we? And I guess that's the whole thing, isn't it? Not everything stays online conveniently waiting to be discovered; things go online, offline. Teams are spread across the world, and that example we shared right at the beginning is a perfect example of that. It just missed the scan time; it wouldn't get detected for 72 hours, and it could go offline and online again depending on how it was activated and set up.
Rachel Schwalk: Yeah, exactly.
Alex Blackman: So that brings us to the end of the webinar; there are no other questions. What I want to do first is just thank everybody for your time in the session today, in particular Andrew and Rachel. I'd encourage you to check out the attachments as well; we've deliberately paired those to go with the webinar. Everything from a dark web risk report tailored to your instance through to preemptive guides that inspired this session today.
If you do have other questions or want to get in touch, please do get in touch with us all on LinkedIn, and follow Searchlight Cyber as well. But otherwise, I just want to thank you again for joining, and say that we hope to see you on a future webinar soon as well. So thank you Rachel, thank you Andrew, and thank you all for joining.
Rachel Schwalk: Thanks.
Andrew Craig: Thank you.










