Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! In July 2026, someone accessed the system Victorian courts use to link participants to online hearings. Not the case management system - the joining list. Four years of it, across ten regional court locations, for the Magistrates' Court and the Children's Court. This episode walks through what was confirmed, what was only claimed, and why a scheduling layer turned out to be more sensitive than the case files it pointed at. What Court Services Victoria confirmed: the incident occurred in July 2026; the data spans 2022 to 2026; ten named regional locations including Bendigo, Castlemaine, Echuca, Kerang, Kyneton, Maryborough, Mildura, Ouyen, Robinvale and Swan Hill. Exposed fields include participant names, case titles and numbers, hearing dates, times and courtrooms, plus two things not on the public record - email addresses and a description of the person's role in the matter. The Case Management System, employee data and financial data were not accessed. What was only claimed: the "more than 28,600 lines" figure comes from the threat actor, not from the organisation. CSV states it "is unable to verify the number of people and matters impacted". Lines are not people, and no affected-person count has been published. The most sensitive claim, carefully sourced: the ABC reports it understands the leaked data includes the names of parties in family violence intervention order hearings and children's court cases. CSV has not confirmed this. CSV's own support page links to The Orange Door, Safe Steps and the Victims of Crime Helpline. Also covered: why nobody has been individually notified, what CSV is offering instead (a public FAQ and a hotline on 03 9087 6116), and why your ability to notify people is a design decision you make years before a breach. Four takeaways: find your scheduling layers; treat metadata and context as data; use retention as a blast-radius control; and know whether your clients sit under the OAIC or a state regulator - CSV is a Victorian public sector body, so this one sits with OVIC under the Privacy and Data Protection Act 2014 (Vic), not the OAIC. And the uncomfortable closing question: CSV confirms this is a completely different system to its 2024 breach. What else in your organisation looks like the system you have not hardened yet? Visit www.kinsoft.com.au to talk through your security and IT needs. Sources: Court Services Victoria, "Data Security Notification" and "CSV data security incident FAQs" (courts.vic.gov.au, 30 July 2026); Magistrates' Court of Victoria notification (mcv.vic.gov.au, 5 August 2026); Cyber Daily exclusive (30 July 2026); Lawyers Weekly (30 July 2026); ABC News (7 August 2026), summarised by DataBreaches.Net; Law360 Australia; Privacy and Data Protection Act 2014 (Vic); Office of the Victorian Information Commissioner.
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.
You know, when you picture a high security digital database, you usually picture the front door. Like you imagine this massive impenetrable steel vault. There are biometric scanners and penetration tests running around the clock. Oh, yeah. Works. Exactly. You've got compliance audits, an entire team of cybersecurity professionals, and their only job is making sure nobody's getting through that front door. And you dedicate, I mean, virtually all of your IT budget to fortifying that specific
entry point. You build the fortress. You document it. You test it relentlessly. But then, and this is the crazy part, you step back and you realize the delivery guy propped open a window in the janitor's closet with a brick just so he wouldn't get locked out. Yeah, the classic back door. Right. And that tiny forgotten window, that's how the entire fortress falls. Oh. So welcome to today's deep dive. We are thrilled you're joining us because we are unpacking a highly revealing cautionary tale from a recent tech analysis by the team at Kinsa. It's a fascinating read. It really is. We're looking at a data breach
that hit Court Services Victoria, or CSV, back in July of 2026. And to really understand the gravity of this, we have to look at what CSV actually does, because it is not a court itself. Right. It's more of the plumbing, almost. Exactly. CSV is the statutory body that operates as the administrative, the IT, and the registry backbone for every single court in Victoria, Australia. the Victorian Civil and Administrative Tribunal. They are the engine room. Wow. OK. Yeah. They
manage the servers. They hold the data. They run the networks. If a judge is viewing a sensitive case file, they're viewing it on CSV infrastructure. Here's where it gets really interesting, though, which brings us to the core irony of this whole event. Hackers did not go after the case files. No. They didn't touch the crown jewel case management system where all the heavily guarded legal records live. They went straight through that propped open window in the janitor's closet. They breached the boring administrative scheduling layer right next to the vault. Right. We're talking about the video hearing invitation system. And the
reason you, the listener, need to pay attention to this is that your workplace, regardless of your industry, almost certainly has one of these invisible systems running right now. Oh, without a doubt. What's fascinating here is the lifecycle of these peripheral systems, mostly because they are born out of, well, sheer panic rather than strategic design. Like during the pandemic. Exactly. Think back to that massive shift to remote work. Organizations needed a way to get people into virtual rooms immediately. They spun up these
scheduling tools purely for convenience. Just to keep the lights on. Right. They were never meant for secure archival storage. They just accumulate sensitive data purely as a byproduct of making the important systems function. Yeah. And then what? Six years later, nobody really owns it. Exactly. Nobody audits it. But it's still running because, you know, turning it off might break a calendar integration somewhere. Which is terrifying. Let's get into the specifics of this breach because the geography and the timeline reveal a lot about how these peripheral systems actually operate in the wild. Yeah. Let's
look at the map. So this attack hit the Magistrates Court of Victoria and the Children's Court of Victoria. But it was heavily concentrated in the regional circuit. Which is a key detail. Yeah. Places across central and northern Victoria. Like looking at the list here, we're talking Bendigo, Castlemaine, Echuca. Krang, Kyneton, Maryborough. It's a massive footprint. It is. Mildura, Uyen, Robinvale, and Swan Hill. It's like 10 different regional hubs. And there's
a very specific, I would say, mechanical reason for that regional concentration. When you're managing court cases across a sprawling geographic footprint like that, the shift to virtual hearings wasn't just a temporary pandemic measure. Right. It became permanent. Yes. It became a permanent logistical necessity. I mean, a magistrate or a specialist lawyer can't lose an entire day driving to Robinville for a 30 -minute procedural hearing. Oh, that makes total sense. So the regional court's dependency on that virtual infrastructure was exponentially higher than in the metropolitan centers, which creates this massive, heavily
concentrated data footprint in this random peripheral scheduling app. Wow. OK, so I was looking at the timeline of how this actually unfolded and there's a gap that feels genuinely concerning. The July timeline. Yeah. So the breach happens in July 2026. On July 15th, a threat actor using the handle of 2019 posts a sample of the data on an underground forum like as proof. And then they dump the entire data set on an anonymous file host. Right. But CSV doesn't publish a public notification or issue a statement to the media
until July 30th. Which seems like a long time. 15 days. 15 days is an eternity in incident response. What were they doing for two weeks while this data was just sitting on the dark web? By August 5th and 7th, the magistrate's court is republishing the notice and the ABC is reporting that police are investigating. It looks slow from the outside, absolutely. But we have to factor in the forensic reality of dealing with unstructured shadow IT. Unstructured meaning like it's not neatly organized. Exactly. Look. When your main SQL database is
breached, your monitoring tools usually scream immediately. You can pull logs, see exactly which tables were queried, and map the whole exfiltration path in a matter of hours. Because it's monitored. Right. But when a peripheral scheduling layer is compromised, you are often dealing with fragmented webhook logs, API calls, and ephemeral data streams. Ah, so it's a mess. A total mess. The security team at CSV likely spent those 15 days painstakingly reverse engineering the attacker's footprints. Just to figure out what this forgotten system
actually contained, let alone securing the vector so the attacker couldn't get back in. OK, that actually makes the media's reaction to this even more interesting because the narrative drifted incredibly fast once that public notification finally did go out on July 30th. It really did. Let's look at what was actually taken. And I will say CSV was. refreshingly transparent here. They listed the breached fields right out of the gate. Which is commendable. Yeah. Now, Australian courts are open by default. So things like case titles, case numbers, hearing dates and times,
courtrooms and participant names, those are technically public record. Right. The critical elements were the two non -public fields that got compromised. The email addresses. Email addresses and the description of the person's role in the matter. Were they the defense lawyer? Yeah. An observer? Court staff? But the vault held. I think that's important to state. There was no employee H .R. data taken, no financial records and most importantly, no substantive court records or transcripts from the case management system were touched. The
core system was safe. Exactly. Yet if you read the mainstream coverage at the time, you would think this was a catastrophic census leak. The hacker, 2019, claimed they had 28 ,600 lines of data. Lines, yes. Right. The media rounded that up to 30 ,000. And within a day, the headline drifted into implying that 30 ,000 people were affected. But CSV completely refused to publish a specific headcount. And I would argue that refusing to publish a number was actually the most scientifically honest thing CSV could have
done in that moment. moment. Hey, I have to push back on that. You're telling me a government statutory body, the IT backbone for entire states judiciary, cannot simply query their own database to count how many distinct individuals were exposed. I'm telling you exactly that. Yes. Don't you just run a deduplication script or something and get your number? See, you're assuming this was a structured relational database built with unique primary keys for every individual. Like a proper user directory. Exactly. But it wasn't. It was an ephemeral scheduling tool. It generates text strings to populate calendar invites. Trying
to run data deduplication on that kind of unstructured export is, well, it's like trying to unbake a cake. Give me a real -world scenario of what that looks like in the data. Like, why is it so messy? Okay, imagine a local legal aid lawyer representing four different clients in four different hearings over a two -week period. Okay, pretty common. That lawyer generates four separate disjointed rows in this export. If a hearing is canceled and rescheduled, that spawns more rows. Oh, I see. If a clerk misspells a participant's email
in one invite but gets it right in another, the system sees two entirely different entities. Right, because there's no central profile checking it. Exactly. And sometimes multiple family members share a single email address for court correspondence. This system was designed to fire off Microsoft Teams links, not to act. as an authoritative census of individuals. That makes a lot of sense. Converting 28 ,600 messy, malformed text strings into an accurate human headcount is technically impossible without making massive assumptions.
So the 28 ,600 number is essentially just a hacker bragging about the size of a text file, not a confirmed victim count. Precisely. But we can't let the technical format of the data distract from the real world implications of what those text strings actually represent. I mean, this isn't a leaked corporate newsletter list. No, it's definitely not. The ABC reported that the magistrate's court hears family violence applications. The children's court handles highly sensitive custodial matters. And this is where we have to fundamentally reevaluate how we think about metadata. We have this dangerous habit in the
tech industry of treating metadata as a lesser harmless class of data. Just administrative exhaust, right? Exactly. A name, a time, an email address. Individually, those are just data points. Yeah. You probably have your name and email sitting on a dozen public websites right now. Sure, yeah. I don't consider my work email address a state secret. But if we connect this to the bigger picture, metadata ceases to be harmless the moment you inject context. Context. Yes. If I take your name and email and I lock it to the context that
you were attending a... regional family violence intervention order hearing on a specific date in a specific role. Wow. Yeah. That is no longer metadata. That is a highly sensitive, genuinely dangerous document. An intervention order exists strictly to keep a violent or dangerous person away from a vulnerable person. The context itself is the payload. Context is data. That is a brilliant way to frame it. And it's clear CSV understood exactly how explosive this context was. Their response showed that, yes. Yeah, when you look
at their incident response, they didn't just throw up a generic IT support page with links on how to monitor your credit score. No, they didn't. They explicitly linked to domestic violence crisis services. The orange door, safe steps, the victims of crime helpline. Which is a heavy thing for an IT department to have to do. Totally. An organization does not put domestic violence hotlines on a cyber breach page unless they know exactly who might be reading it and that their physical safety is actively in jeopardy. It demonstrates a very mature understanding of their threat model.
They recognize that the risk wasn't financial fraud. It was physical stalking and retaliation. But knowing that severe human risk, we have to critically evaluate how CSV actually responded to the crisis. It's a mixed bag, structurally speaking. Right. So technical containment is one thing. They shut down the access vector. They confirmed the main case management system was untouched, which helps stop the media speculation. This is good. But they made a very controversial decision regarding the victims themselves. They chose not to directly notify the individuals
whose data was leaked. Yeah, that caused an uproar. They just set up a public FAQ and a dedicated phone line. It was Air 3, 9087 -6111. But there were no direct emails sent out to the affected parties. And that decision drew a lot of criticism. But the analysis from Kinsoft defends it as an impossible choice. OK, so what does this all mean? Are they just taking the easy way out by not emailing the victims? Because if my involvement in a family violence hearing was just dumped on the dark web, I would want to know immediately.
Of course you would. Relying on victims to proactively check a public FAQ and dial a hotline. It feels like the organization is washing its hands of the hardest part of incident response. It feels like an abdication of responsibility. Absolutely. But let's connect this back to what we just discussed about the unstructured, messy reality of the data. Cheat text strings. Yes. CSV couldn't accurately deduplicate the rows. If they attempted a mass email notification based on unverified, fragmented text strings, the collateral damage would be immense. How so? I mean, if you email the wrong
person, they just ignore it, right? It goes to spam. Not in this context. Imagine a scenario. Where an abusive ex -partner has access to a shared family email address or compromised the victim's email account years ago. Oh, wow. If CSV sends an automated breach notification to that address saying your attendance at a family violence hearing was exposed, the notification itself acts as a trigger for violence. Oh, my God. I didn't even think of that. Or what if the scheduling tool had the wrong email entirely?
and a complete stranger is suddenly informed about a neighbor's highly sensitive family court matter. When your data quality is that poor, the act of notification becomes a severe security risk in itself. That completely flips the paradigm. The inability to safely notify people isn't a failure of the crisis response team. It's a failure of the system's architecture from day one. Precisely. Because they built a quick scheduling tool instead of a secure identity platform, they permanently stripped themselves of the ability to safely warn victims years later. Yes. Your crisis options
are dictated entirely by the design decisions you made during procurement. Which naturally leads to the broader organizational takeaways here. The Kinsoft source breaks down four key lessons for anyone listening, starting with find the scheduling layers. We have to start actively hunting for these hidden layers in our own networks. You have to ask yourself, what did your workplace spin up in 2020 that is still running? Right. A visitor management iPad in the lobby. A telehealth joining link. A room booking app. You have to look at those systems and ask, if this entire
database was published on the Internet tomorrow, what would it reveal? Which leads to lesson two. Context? is data. Exactly. As we established, merely recording that an event happened is a massive disclosure. If your company uses software to track who books the internal HR grievance room, that schedule is highly confidential data, regardless of whether the actual meeting notes are secure. And that ties perfectly into lesson three, which is frankly the most frustrating aspect of this entire breach. Retention is security. The data CSV lost spanned from 2022 to 2026.
Four years of data. Four years. They were holding on a temporary video link scaffolding for four years. That drives me crazy. What possible operational purpose is served by keeping a meeting link for a procedural hearing that concluded in 2022? None. There is none. The case is closed. The legal transcripts are locked away in the secure vault. This scheduling data is essentially... digital exhaust and they were just hoarding it. We treat data storage like a digital junk drawer because cloud storage is practically free. But
hoarding ephemeral data is just hoarding risk. Retention policies are the cheapest, most effective security control any organization possesses. You just said it to delete automatically, right? You simply configure the system to automatically delete administrative logs after 60 or 90 days. If CSV had implemented an automatic 12 -month deletion rule on that peripheral app, they would have reduced the blast radius of this hack by 75%. For zero cost. Well, nobody ever bothers to turn those controls on for the boring administrative apps. And that oversight gets heavily compounded
by Lesson 4. Regulatory blind spots. Oh, this is a big one. It is. When most IT professionals hear data breach in Australia, their minds immediately jump to federal legislation. They start pulling up playbooks for the Privacy Act and the Office of the Australian Information Commissioner, the OAIC. But CSV is not a federal body. They are a state agency. Exactly. They fall under the Victorian Privacy and Data Protection Act of 2014, which means they are regulated by the Office of the Victorian Information Commissioner, OVIC.
Let's pause on that because the distinction between OVC and OAIC sounds like, you know, bureaucratic alphabet soup. It does. But the implications for an incident response team are massive. Thousands of private IT firms provide services to state government clients. If their internal crisis runbook is only built for federal law, they are essentially playing chess with checkers rules. That is a perfect analogy. Victoria, New South Wales, Queensland and the Commonwealth all operate under entirely different notifications. schemes, different thresholds for harm, and different
regulatory bodies. So you can't just have one playbook. No. If your managed service provider's runbook mandates notifying the OAIC within 30 days, but your client is a Victorian state agency with a fundamentally different statutory obligation, your incident response plan is legally noncompliant before you even execute step one. Wow. IT vendors have to map these regulatory boundaries today, not while the network is actively bleeding data. It really is a masterclass in modern risk management.
You have to map your hidden layers, recognize that context transforms metadata into a weapon, weaponize data retention as a fence mechanism, and clearly define your regulatory jurisdiction. That covers all the bases. But as we wrapped up this deep dive, there is one final detail from the source material that ties this entire narrative together. Down in the FAQ that CSV published, they explicitly confirmed that this This compromised scheduling system is completely unrelated to a previous highly publicized breach they suffered back in 2024. It is a very truthful clarification, but it also accidentally summarizes
the core failure of modern cybersecurity. Well, after CSV had their severe incident in 2024, they did what any rational organization would do. They poured resources into heavily fortifying the specific system that got breached. They bought a thicker, heavier vault door. They added more lasers and more guards to the front entrance. Yes. But threat actors do not care about your new vault door. They aren't fixated on a system just because you recently hardened it. They are attacking the entire organizational surface area.
So they just walk around the building looking for a window. Exactly. The next breach never comes through the door you just reinforced. It comes through the one you completely forgot was even a door. When an organization finishes a post -incident review, the question shouldn't just be, did we fix the thing that broke? The essential question is, what else in our tech stack operates exactly like the broken thing did before we realized it was vulnerable? That is a fundamental shift in how we need to view network architecture. It's not about defending the castle. It's about knowing where all the
tunnels are. And if we step back and look at the broader trajectory of our digital infrastructure, this incident forces us to confront a very uncomfortable reality about shadow IT and convenience. Okay, I'm listening. Think about it. If every major organization, governments, hospitals, corporations, is legally fortifying their centralized legacy databases. The vaults. The vaults. But simultaneously, they are offloading the messy logistical work to dozens of third -party integrations, temporary
microservices, and unvetted scheduling apps just to make daily operations frictionless. Well. We aren't actually securing our data. We're just moving the risk around. Worse. We are simply building a society where our most sensitive human moments are perpetually scattered across a forgotten, invisible periphery. We have to ask ourselves, when our fragmented digital identities are permanently accumulating in these neglected administrative layers, is the illusion of a secure vault just a comforting lie we tell ourselves to keep the calendar invites flowing? Are we trading our
SASE just for the sake of seamless digital convenience? That is a haunting thought to leave on. The front door might be locked tight, but we are the ones propping the windows open just to make the day run a little smoother. Thank you so much for joining us on this deep dive. Take a hard look at the invisible doors in your own digital lives, and we'll catch you next time.
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...
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...
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...
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...
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;...