Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Tech Talks With Kinsoft

Tech Talks With Kinsoft artwork

DHS HSIN – When "Unclassified" Doesn't Mean "Low Risk"

Jul 9, 202624 minAudio

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! A breach at the heart of America's homeland-security apparatus - and a masterclass in why "unclassified" is not the same as "unimportant." The US Department of Homeland Security has confirmed that attackers broke into the Homeland Security Information Network, or HSIN - the platform it uses to share sensitive-but-unclassified information with thousands of federal, state, local and private-sector partners, and one currently helping coordinate security for the 2026 World Cup. Classified systems were untouched, but what the intruders took, and who they are, remains unknown. We dig into the danger of "sensitive-but-unclassified" data, why legacy collaboration systems are prime targets, and how to communicate honestly when you know attackers got in but not yet what they took. Do you know how your own data is classified - and protected accordingly? Visit www.kinsoft.com.au to talk through your security and IT needs. Sources: BleepingComputer; Nextgov/FCW; TechCrunch; PYMNTS; UpGuard.

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show. Tap a time to play from there.

Imagine for a second leaving the master blueprints, the guard shift schedules, the camera blind spots and even the emergency evacuation plans for the 2026 World Cup. Just, you know, sitting unattended on a park bench. Just completely out in the open. Right. Completely out in the open. Anybody walking by could just pick them up, read them and walk away without you. ever knowing it happened yeah which is a terrifying thought it really is and today we're looking at a cyber security incident from july 1st 2026 where the u .s department

of homeland security effectively realized someone might have done exactly that but you know in the digital world exactly so welcome to the deep dive The source material for this discussion is a major security briefing covering a breach of the DHS information network. And our mission today is to, well, separate the confirmed facts from the rampant rumors that are circling Washington right now. And there are a lot of rumors. Oh, tons of them. But most importantly, we're unpacking a central, incredibly dangerous assumption that

organizations make every single day. We're looking at what happens when an organization confuses the word unclassified with the word unimportant. Yeah. And the distinction, it really is the beating heart of this entire incident. I mean, what makes this briefing so compelling to me isn't it's not the technical sophistication of the malware they use. Right. It's not some crazy zero day exploit or whatever. Not at all. It's the realization that the data we just assume is safe simply because it doesn't carry this big red top secret stamp on it. The data can actually hold the keys to

the kingdom. We're looking at a fundamental misunderstanding of how data actually gains value in the modern era. Okay, let's unpack this. Let's start by mapping out the target itself. So the network breached here, it's called HSIN. That's the Homeland Security Information Network. And I got to say, when I first read that, I pictured like a generic internet site. Like a basic employee portal. Yeah, exactly. A place where government employees go to read HR policies, download tax forms, figure out what days they have off, that kind of thing. But looking at the briefing, this is something

entirely different, isn't it? Oh, it's far different. I mean, HSIN is essentially the primary information artery for the Department of Homeland Security. Wow. OK. Yeah. It's the main platform they use to communicate across every single level of governance and security in the United States. Wait, every level? Pretty much. We are talking about federal agencies, state and local law enforcement, tribal governments, territorial governments, and even private sector organizations that manage critical infrastructure. Okay, so if I'm trying to visualize

this, it's basically the ultimate behind -the -scenes group chat for every level of security and law enforcement in the entire country. That is actually a really good way to put it. So if, like... a local police department in Dallas needs to coordinate a threat response with a federal agency, or if stadium security at some major venue needs to pass a tip up to the feds, HSIN is the room where that happens. It's the central nervous system for keeping things safe. Right. And because it functions as that central nervous system, the operational tempo of the data flowing

through it is just intense. I mean, they don't use this for boring press releases. No. No, they use it to coordinate security for massive planned events. They rely on it for active incident response management when a crisis is unfolding in real time. Jesus. Yeah, it's the main repository for sensitive ongoing threat assessments and, you know, information regarding active persons of interest. Okay, wait, hold on. If it's that sensitive, I'm actually slightly hung up on the mechanics here. Sure, what do you mean? Well, why use a massive sprawling network that invites so many

different people in if you're sharing active threat data? Why not just use secure encrypted channels between very specific agencies when they actually need to talk? Well, because scale and speed dictate the architecture here. I mean, in a crisis or when you're coordinating a nationwide event, you don't have the luxury of trying to figure out which of the, I think there's something like 18 ,000 local law enforcement agencies in the U .S. Wow, 18 ,000. Yeah, a huge number. You can't figure out which of them need to be on some specific encrypted email chain. Okay,

that makes sense. You just don't have the time. Exactly. You need a centralized platform where vetted partners can just log in. access real -time dashboards and share intelligence instantly. The sheer volume of coordination requires a platform of this size. Right. But of course, that massive footprint is exactly what makes securing it such a monumental challenge. Which brings us to the intrusion itself, the part everyone is talking about. Let's look at the timeline laid out in the reporting. Yeah, the timeline is interesting. It is because the tech outlet NextGov actually

broke this story before the government formally announced it. So the intrusion is believed to have happened sometime between late May and early June of 2020. And then DHS came out on July 1st and officially confirmed, yes, we had a cyber incident. Right. They had to own up to it. Yeah. And the attackers successfully compromised the HSIN servers, specifically targeting an associated legacy SharePoint system that was used for internal collaboration. And I will say the official response was swift. Once the breach was discovered, they

didn't just sit on their hands. No, they didn't. DHS confirmed the incident. They isolated the affected systems immediately to prevent the infection from spreading. Yeah. They mitigated this specific vulnerability the attackers used. Right. And they launched a forensic investigation right away. They also went out of their way to explicitly state that their classified networks, the systems holding actual hard state secrets, were completely untouched. Untouched. OK. But reading the briefing, it highlights this massive, glaring unknown. We know the attackers got in. DHS literally admitted

the doors were breached. Yeah, they were inside. Yet, as of right now, nobody actually knows what, if anything, was taken. And I have to ask. How is that even possible? How can a modern tech team know someone was wandering around inside a server but have no idea if they walked out with armfuls of files? I know it sounds crazy, but it really comes down to the difference between access logs and exfiltration logs. Okay, break that down for me. So in cybersecurity, we actually frequently encounter these access -confirmed, impact -unknown scenarios. When investigators

look at the digital footprints left behind, they might find a compromised credential or backdoor installed on a server. Right, like finding muddy boots in the hallway. Exactly. That proves the attacker was standing in the room. But unless your network is specifically configured to monitor and log the outgoing flow of data... Like tracking what leaves the building. Yes, tracking exactly which files were copied and sent outside the network. Unless you have that, you cannot definitively prove... what they packed into their briefcase before leaving. Wow. So they just don't have

the cameras pointing at the exit, so to speak. Basically, yeah. Some systems just aren't instrumented to record that level of outbound traffic detail. And that is especially true for older legacy systems, which this was. I see. And honestly, that leads to the other really strange element of the story. Here's where it gets really interesting to me. Usually when we cover massive breaches, there's this, I don't know, this. theatrical element to it. Oh, absolutely. Our ransomware gang steals 50 million records. They go on the

dark web and they shout from the rooftops, hey, pay us 50 million dollars or we leak everything tomorrow. Right. They want attention. But here there is total silence. No threat actor has claimed responsibility at all. DHS hasn't attributed it to a criminal syndicate or a foreign government. Doesn't that lack of a boastful threat actor actually make this significantly scarier? Oh, it changes the entire threat model. completely how so well when a ransomware gang makes a bunch of noise their motive is transparent right it's extortion they just want to pay out yeah it's

just about the money exactly silence on the other hand almost always points to espionage it points to a highly sophisticated threat actor very often a nation state, who wants to establish persistence in the network. They want to live there. Yes. They want to quietly slip in, set up camp, and silently siphon off intelligence for months or even years without triggering any alarms whatsoever. Yeah, and that's chilling. It is. The value of the data to an espionage unit is in secretly possessing it. not in selling it back or bragging

about it on Twitter. So the fact that DHS knows a silent actor was inside but doesn't know what they took. It is creating an absolute vacuum of security confidence in Washington right now. And looking at the stakes, that alarm makes total sense. Because we aren't talking about stolen credit card numbers or embarrassing emails. The briefing notes that a senior U .S. senator, actually the vice chair of the Senate Intelligence Committee, publicly demanded a Justice Department investigation citing a severe national security risk. Yes. And he didn't mince words. No, he didn't. And

the context here is super heavy. HSIN is currently one of the primary platforms being used to coordinate security for the 2026 FIFA World Cup, which is obviously being hosted right here across multiple cities in the United States. Yeah, the immediate. Real world application of that data, the physical reality of it, it just cannot be overstated. We are talking about the tactical reality on the ground for one of the most high profile gatherings on the planet. Millions of people. Millions of fans, global dignitaries, sprawling infrastructure

across dozens of jurisdictions. Right. Which goes right back to my opening thought. It's not a leak of mere marketing data. If an adversary gets their hands on this, they are looking at the patrol routes of security personnel. Yes. They are looking at the response times for emergency medical teams, the known camera blind spots in a stadium. It's operational security information. And if someone knows exactly how you plan to respond to an incident, they can just reverse engineer a way to defight that response. Which is exactly why this pivots the discussion to a much broader issue. And honestly, one that

affects every single organization out there, not just government agencies. OK, let's get into that. The data sitting in that breached SharePoint server was. By definition, unclassified. It did not carry a secret or top secret clearance level. OK, I have to stop you there because this is where my brain just gets totally stuck. Go ahead. If the data includes the security blind spots and patrol routes for the World Cup, how on earth is that not classified? Why wouldn't they stamp top secret on that instantly? I get why you'd

think that, but it's because of the incredibly rigid criteria of government classification. Really? Yeah. Classified information usually relates to things like intelligence sources, undercover operatives, military operations, or foreign relations, where unauthorized disclosure could cause exceptionally grave damage to national security. Like nuclear codes and spy identities. Exactly. A local police department's shift schedule for traffic control outside a stadium simply does not meet that threshold. An email discussing

the location of medical tents does not meet that threshold. I mean, I guess that makes sense. Individually, these documents are entirely mundane. So by definition, they are unclassified. Individually. I see where you're going. Yes. And this is the trap of labels. Humans have this psychological tendency to view anything labeled unclassified or internal as low value. But unclassified definitely does not mean unimportant. Right. And the briefing emphasizes this concept of the aggregation effect.

Yes. Aggregation is the key here. Let me see if I have this right. I like to think of it kind of like a massive jigsaw puzzle. OK. Like that analogy. So if I hand you one single cardboard puzzle piece and it's just a blob of blue color, it tells you absolutely nothing. It is unclassified. It is unimportant. Right. It's just blue cardboard. But if I managed to sweep up 10 ,000 of those blue pieces. and snap them all together. Suddenly, they reveal a highly detailed blueprint of the entire sky, or in this case, the entire security apparatus. That captures the mechanics of aggregation

perfectly. A sophisticated threat actor isn't usually looking for one dramatic spy dossier like in a James Bond movie. Right. They don't need the one glowing floppy disk. Exactly. They want all the little pieces that build the context. When you aggregate thousands of unremarkable, unclassified documents, things like event security plans, agency contact details, localized threat assessments across 30 different cities. They compound. They add up. They add up to something extraordinarily sensitive. They create a comprehensive map of the systemic capabilities and vulnerabilities

of the Department of Homeland Security. And the briefing notes that technology has made this easier than ever for the attackers to actually do. I mean, it's not just a room full of spies reading through PDFs one by one anymore, is it? Oh, not at all. Modern threat actors leverage machine learning and AI to ingest and process massive amounts of stolen data instantly. So they don't even have to read it. Nope. They can scrape a compromised SharePoint server, pull out millions of seemingly boring emails and schedules, and just use automated tools to map relationships, identify key personnel, and highlight systemic

security gaps. Wow. Yeah. The AI does the puzzle assembly for them in a matter of seconds. So what does this all mean for the listener? Because bringing this down to earth... Everyday businesses must fall into this exact same trap all the time. Constantly. I mean, if we connect this to the bigger picture, think about the architecture of a typical corporation today. Okay. They will spend millions and millions of dollars building digital fortresses around what they consider obviously secret. Like the proprietary source

code. Right. The source code, the unreleased financial reports, the CEO's private communications. They lock those down tight. As they should. Absolutely. But then they leave everything labeled general or internal wide open to anyone with a basic company login. You mean things like the employee directories, the meeting minutes, onboarding manuals? Internet pages showing how the IT help desk operates. Yeah, all that mundane stuff. All of it. Organizations forget that if an attacker gains access to a thousand mundane internal documents,

they can map out the entire corporate structure. They can see that, say, an entry -level employee in marketing. has access to the same shared drive as the VP of Finance. Oh, that's dangerous. Or they can read the IT manual to learn exactly what software the company uses for remote access. Which is basically handing the attackers the script for a devastating social engineering attack. Exactly. Like, if I want to spearfish your company, I don't need your source code. I just need to read your internal newsletter so I know your IT director is named Susan, she's on vacation

this week, and the company just switched to a new payroll vendor. Yep, you have the context now. I use that mundane, unclassified info to craft an email that looks 100 % legitimate. And that is exactly how the fortress falls. Businesses make this same mistake in miniature all the time. The solution isn't to lock down every single file with biometric security, though. I mean, that would grind business to a halt. Right. You still have to get work done. The solution is to evaluate your data based on the impact it would have in aggregate if it leaked, not by

whatever comfortable legacy label it currently has. You have to protect the merely sensitive data with real, robust access controls. Okay, so if this aggregated data is such a goldmine, where is it usually sitting? Because this brings us to the second big lesson and a crucial detail in the DHS breach. The legacy systems. Exactly. The attackers didn't hit a shiny new database. The briefing specifically highlights that they targeted a legacy SharePoint system used for internal collaboration. Yeah, the threat of legacy platforms is a massive blind spot for almost

everyone. When we say collaboration platforms, we're talking about internal wikis, shared network drives, old SharePoint servers, intranets. These are environments designed to be highly accessible so people can just dump files, share drafts, work together easily. Right. They are kind of the digital equivalent of a company attic. It's where half -finished projects from five years ago live, where old contact lists sit gathering dust, where drafts of operational plans are just completely abandoned. And organizational knowledge

quietly accumulates there for years and years. The problem is, defenders routinely overlook these systems. Why is that? Because IT departments naturally focus their security budgets and their monitoring tools on the internet -facing systems. The public stuff. Exactly. The customer portals, the public websites, the payment gateways. They secure the front door with everything they have, but they completely forget about the sprawling, messy internal shared drive sitting on a server in the back room. I have to ask the obvious question,

though. If these legacy systems are such a huge risk, why don't companies just delete them? Why doesn't IT just hit the kill switch on a SharePoint server from 2018? Because it is rarely that simple. Really? Just unplug it. I know. But first, you have complex data retention laws. Certain industries are legally required to keep communications and operational data for up to a decade. Okay. Fair enough. And second, there is a massive institutional fear of breaking things. Oh, I've seen that. Right. What if a critical accounting macro that

runs payroll is quietly referencing some obscure file on that old server? If IT deletes the server, payroll crashes and nobody knows how to fix it because the person who built it left the company three years ago. Ah, yes. The classic it's running fine, please don't touch it mentality. Exactly. IT departments are notoriously overworked. So if a legacy system is running quietly in the background and not causing immediate downtime, the incentive is to just let it be. But the result is that these systems become older, less monitored

and unpatched. Yes. And to an attacker, an unpatched, unmonitored legacy server packed with a decade of aggregated internal data. I mean, it's the absolute holy grail. They look for these forgotten environments specifically because they know nobody is watching the logs. So anyone listening right now who manages IT infrastructure needs to realize that an on -premises legacy server requires the exact same aggressive patching urgency. as their public -facing website, because the attackers see those invisible systems perfectly clearly.

Without a doubt. They're heavily targeted. Well, we have established a pretty grim picture here. The trap of unclassified labels, the danger of aggregated data, the vulnerability of forgotten legacy systems. But the source material does offer a silver lining, right? It does. There is one major thing that DHS actually got right in this scenario, which is why this breach was contained rather than becoming a total national catastrophe. Saving Grace is network segmentation. Lesson number three. Let's break down how that

actually works, because it sounds a bit like corporate jargon. Sure. According to the briefing, DHS had strictly separated their classified networks from their unclassified network. So even though the attackers got into the legacy SharePoint system, they couldn't reach the crown jewels. I visualized this like the watertight doors on a submarine. Oh, that's a great visual. Yeah, like if a torpedo breaches one compartment, water rushes in. But the crew instantly seals the heavy steel dholes around that section. So you lose that specific compartment, but the whole sub

doesn't sink. The submarine analogy is highly accurate for how modern network architecture should be designed. In cybersecurity terminology, we call this limiting the blast radius. Limiting the blast radius. So how does that actually function on a technical level? What are the watertight doors in a computer network? It involves physical and logical separation. On a basic level, it means setting up strict access controls, firewalls, and separate subnets for different departments or data classification levels. So just because an employee has valid credentials to log into

the general HR portal, that does not mean their computer is physically or digitally allowed to route traffic to the server holding proprietary source code. The network simply drops the connection. So it prevents what they call lateral movement. Yes, exactly. When an attacker breaches a network, say they phish an entry -level employee, They usually land on a low -level machine. Their very next step is lateral movement. They try to spread out. They try to crawl through the network, escalating their privileges, searching for those high -value databases. If the network is flat, meaning everything is just connected to everything else without

internal barriers, they can just walk right up to the crown jewels. Basically a submarine with no internal doors. One leak sinks the whole ship. Precisely. But with strict segmentation, the attacker hits a firewall. They try to move from the unclassified SharePoint server to the classified intelligence database. And the network demands a completely different set of credentials, physical tokens, or even requires them to be on a completely separate, physically disconnected network. They are trapped in the compartment. Exactly. DHS survived this specific intrusion because their

watertight doors held. A breach of the unclassified network did not bleed over into the classified secrets. This feels like a direct challenge to anyone listening right now. Think about it. If a hacker got into your general environment tomorrow, if they compromised a single laptop in your marketing department, what is actually stopping them from walking straight into your financial databases or your customer records? It's a question every IT leader needs to ask. Right, because if your network is just one big open floor plan, you have a massive vulnerability waiting to be exploited.

Segmentation acknowledges a very harsh reality of modern security, which is that... breaches are going to happen. It's inevitable. Sophisticated attackers will eventually find a way over the outer wall. So the goal isn't just to build an impenetrable perimeter. The goal is to ensure that when they do get over the wall, they were trapped in the courtyard and cannot breach the castle keep. It's about building resilience into the architecture. So as we synthesize all of this, the narrative of the DHS breach really comes down to three operational imperatives that apply to almost any environment. The takeaways

are incredibly clear here. First, You must recognize that aggregated, unclassified information is a goldmine. You have to evaluate your data based on the sum of its parts and the impact of a leak, not just by its individual, comfortable label. Second, you cannot ignore your legacy systems. The digital attic. Yes. Those collaboration platforms are treasure troves of institutional knowledge that demand aggressive patching, monitoring, and eventual decommissioning when they outlive

their utility. And third, implement strict network segmentation. Build those watertight doors so that a single compromised email account doesn't bring down the entire organization. Because ultimately, attackers do not respect the labels you assign to your data. They only respect access, and they capitalize on accumulation. Which leaves us with a final thought to mull over today. We've spent this entire deep dive examining massive organizations, government intelligence, and the coordination of the World Cup. But I want you to bring these concepts down to a deeply personal level. Oh,

this is a great point. Think about your own digital footprint. Think about the aggregation of your lifetime online. The personal legacy data. Exactly. If unclassified organizational data becomes highly sensitive and dangerous when an attacker aggregates it, Consider how much of your own scattered, seemingly unimportant personal data is just floating out there. A staggering amount for most people. Right. The abandoned social media accounts you haven't checked since college. The old emails from a decade ago just sitting in a forgotten inbox. The legacy cloud drives, digital receipts,

location check -ins on apps you don't even use anymore. Yeah, it adds up fast. It really does. If a bad actor simply swept all of those mundane fragments up into one big pile and fed them into an AI, what kind of vivid, highly compromising picture would it paint of your life? A very detailed one. It really makes you realize that whether we are talking about a massive government security plan or just your digital grocery lists, there is a massive difference between unclassified and unimportant. A profound difference. Yeah. And it is a distinction that every organization

and frankly, every individual needs to take far more seriously in the years to come. Absolutely. Well, thank you so much for joining us for this deep dive. Stay patched, stay skeptical, and we will catch you next time.

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. Iris Ohyama - A Suspicious File in the Cloud, 7.7 Million Customers at Risk, and an Online Store That Went Dark

    Nov 19, 202620 minAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! Japanese home goods company Iris Ohyama (online store operator Iris Plaza) found a suspicious file in an external...

    Credit the people on this episode

  2. Who Gives A Crap - A Marketing Supplier Breached, Six Days to Tell the Client, and Customer Data That Didn't Need to Be There

    Nov 17, 202619 minAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! Australian online retailer Who Gives A Crap told customers on 7 November 2026 that their information was accessed...

    Credit the people on this episode

  3. Last Week in Tech - An Outlook Zero-Day That Leaks Your Password Hash, Chrome's Eighth Exploited Bug, and Nvidia Without Jensen Huang

    Nov 15, 202621 minAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! Last week in tech, 7 to 14 November 2026. Microsoft's November Patch Tuesday (10 November) fixes well over a...

    Credit the people on this episode

  4. Chainalysis - From "Non-Sensitive" to "All Users": How a Breach Notice Changed in Five Days

    Nov 12, 202619 minAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! Chainalysis disclosed on 30 October 2026 that an external actor using "offensive agentic techniques" accessed...

    Credit the people on this episode

  5. NSW Health - A Dental Voucher Website, Four Gigabytes Taken, and the Providers Whose Details Were Exposed

    Nov 10, 202623 minAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! NSW Health reported unauthorised access to its Oral Health Fee For Service Scheme website on 29 October 2026;...

    Credit the people on this episode