Beyond the Checkpoint: Borders as Strategic National Assets

Good morning everyone and thank you for the invitation to provide an address at this summit. It’s a privilege to be here with so many colleagues from across government, industry, and the technology sector.

I want to begin by paying tribute to those who are on the front lines – the ABF officers, AFP agents, biosecurity inspectors, and the many others who manage Australia’s border every day. Whatever we articulate in rooms like this, it’s their professionalism and dedication that ultimately delivers border security. Any technology we develop must be designed to help them do their job, not replace their judgment.

Now, I want to start somewhere personal. The last time many of you heard me speak publicly was at the National Press Club, in October 2024, where I delivered what was effectively a valedictory address after 44 years in law enforcement and border management. I said at the time that it was the right moment to close the book, quench the candle and ring the bell on my career.

Well, it turns out retirement is harder than it looks.

I tried golf. I got tennis elbow.

I’m told this is not uncommon for people who confuse enthusiasm with technique. Three months of so-called rest and I was climbing the walls.

But in those first months, something more serious happened. I found myself unable to stop thinking about borders. Not operationally – I was confident the new Commissioner had that well in hand. But at a national level. The more I read about what was happening globally – the acceleration of US-China competition, the weaponisation of supply chains, the convergence of state and criminal threats – the more I felt that the arguments I’d been making as Commissioner needed to be elevated, made more loudly, and from a different vantage point.

I decided that I’d initially go it alone as a sole trader – as an independent advisor to industry. Which, I should tell you, is not as straightforward as it sounds — there are rules around post-separation employment for former senior officials, and navigating them carefully has been important to me.

I also started writing what became an Occasional Paper for the ANU’s National Security College, which was published last month – Five hundred hours of research and writing later, I came to understand why the College calls them ‘Occasional Papers’.

That paper, “Beyond the checkpoint: Managing Australia’s border as a strategic economic and national security asset” is the foundation of what I want to talk about today. It’s my own work, grounded in my own experience, and it represents the arguments I want to present about the Australian border. Why beyond the checkpoint? Put simply, in my experience most people, including in government, have a very narrow conception of a “border” — based on their personal interactions with it — a smart gate, passenger card or customs declaration.

Today I’m speaking in my capacity as an independent advisor. I’m not representing any company, agency, or interest. I’m making a case – one I believe in deeply – that Australia’s border should be understood not as an operational group of functions, or a regulatory checkpoint, but as a strategic national asset. And that the technology we discuss in this room is central to whether that asset can deliver both the assurance and throughput that are necessary for our security and prosperity, in a more volatile world.

Part 1: Deglobalisation and US-China Strategic Competition

Let me start with the strategic context, because technology choices don’t happen in a vacuum.

The assumptions that underpinned decades of border management – open markets, rules-based trade, predictable supply chains – are fracturing. Since 2018, thousands of new trade restrictions have been imposed globally, nearly 3000 in 2023 alone. Since then, the United States has implemented sweeping tariffs, export controls on critical technologies, and investment screening mechanisms that would have been unthinkable a decade ago. Meanwhile China has implemented its own controls on critical minerals and strategic commodities, whilst dramatically expanding its Belt and Road initiative and its component Digital Silk Road standard setting apparatus.

These aren’t discreet trade disputes, but a fundamental restructuring of the global economic order. And for Australia – a trade-dependent nation, deeply integrated into Indo-Pacific supply chains, caught between our largest trading partner and our principal security ally – the implications are profound.

We had an early experience of what this world looks like. When China imposed restrictions on Australian coal, wine, barley and other commodities from 2020, we had to rapidly understand alternative supply chains, identify new markets, and monitor compliance with evolving restrictions. When the US implemented export controls on advanced semiconductors, Australian companies suddenly faced new compliance obligations that rippled through their supply chains. AUKUS, Free Trade Agreements, and Comprehensive Strategic Partnerships add yet more complexity into the mix.

The UK’s Integrated Review Refresh warns of trade, technology, and supply chains being used as tools of statecraft. Japan’s National Security Strategy identifies economic coercion and supply chain fragility as major security challenges. The Europeans are building their own economic security doctrine. Everyone is responding to the same reality: that the era of frictionless globalisation is over, and borders have become contested spaces where economic and national security converge.

So, what does this mean for the border systems we build? It means they must provide strategic visibility – the ability to see, understand, and act on what, and who, is moving across our border in near real time. When sanctions or trade restrictions are imposed overnight, our systems need to identify affected goods and entities immediately. When supply chains are disrupted, we need to understand the second and third-order effects. When critical inputs become strategic leverage, we need to know our dependencies before they’re exploited. This is not the border technology paradigm we have inherited. But it’s the one we need.

Part 2: Criminal Exploitation and System Vulnerabilities

The geopolitical shifts I’ve described haven’t just created new compliance pressures. They’ve created new opportunities for criminal exploitation. And those opportunities are being seized with alarming sophistication.

Transnational serious organised crime has never been more networked, or more deeply embedded in legitimate systems.

When I established Operation Jardena following the AFP’s Operation Ironside, we identified around 100 companies and 1,000 individuals operating at the Australian border — in our air and seaports, across our supply chains — with direct or close links to organised crime groups. There is similar infiltration in many overseas ports and supply chains.

That wasn’t news to those of us working the problem. But the scale was confronting. And it confirmed what we’d long suspected: the border is seriously infiltrated, and traditional enforcement models cannot address it at scale.

Organised crime doesn’t just smuggle contraband past the border. It embeds itself within the border. Corrupting workers, establishing front companies, exploiting insider access. These networks are persistent, sometimes building legitimate-looking businesses before using them to move contraband, recruiting and cultivating workers who sometimes may not know who they’re ultimately working for. And they have deep pockets.

The economics are stark. In Australia alone, the Illicit Tobacco and E-Cigarette Commissioner estimates the combined illicit tobacco and e-cigarette market is worth between AUD 5.7 and 8.5 billion annually, with foregone excise revenue of up to AUD 11.8 billion. That makes illicit tobacco and vaping among the country’s largest single sources of criminal proceeds.

Globally, the proceeds of organised crime are estimated at around 1.5% of global GDP — around USD 1.76 trillion in 2025. AUSTRAC estimates roughly half is laundered through the financial system. Where’s the rest? Much of it is almost certainly moving through trade — through phantom shipments, falsified quantities, and manipulated invoices that deliberately misstate the value of goods crossing borders. It’s called trade-based money laundering, and between 2008 and 2017 alone, this kind of invoice fraud across developing countries was estimated at USD 8.7 trillion. It is one of the great blind spots in global financial crime — and our current regulatory architecture simply cannot see it.

And here’s where the border’s challenge is evolving in ways that many people haven’t yet grasped.

Today’s threat actors operate across domains that no single agency owns — including digital payment channels that can bypass the correspondent banking system entirely, leaving no suspicious transaction report, no funds transfer instruction, no sanctions screening trigger.

Consider illicit tobacco. It is ordered online, paid for in cryptocurrency, and shipped under falsified manifests — sometimes all three in the same transaction. Each element of that supply chain leaves a digital trace that simply didn’t exist a decade ago. Cryptocurrency transactions, supply chain sensors, financial flows, customs declarations — the data is there. What’s absent is the architecture to fuse it lawfully and at speed.

This is a regulatory gap, not a technology gap. The same pattern holds whether were talking about illicit commodities, terrorist financing, sanctions evasion or trade based money laundering disguised as legitimate trade. The transaction looks different each time. The blind spot is the same. AI gives us the capacity to connect these traces across datasets and see networks rather than isolated seizures — but only if our institutional arrangements are built to allow it. Right now, too much of that analytical capability sits with a small number of specialists. The opportunity is to embed it across the system.

And here’s a structural constraint that frames everything else. Historical audits show Australia once examined around 7% of sea cargo containers. Today, with dramatic growth in containerised trade, coverage has fallen below 1%. This is not a criticism of our border officers — on the contrary they do extraordinary work. But they are constrained by an operating model that relies on the physical inspection of selected goods at the point of arrival, based on intelligence-led/risk-based targeting. This model produces low coverage and strike rates. It’s clear that this operating model will be unable to deliver the level of assurance our community reasonably expects going forward — most threats are concealed within the 99% of goods we don’t examine.

The Checkpoint Model Is Exhausted

Technology is the only way to change this equation. Not as an add-on to existing processes, but as the foundation of a fundamentally different operating model at the border.

It starts with digital visibility — fusing trade data, financial, and law enforcement holdings so that we can see risk before it arrives at the physical frontier. That visibility will sharpen precision — the ability to target the right container, the right shipment, the right entity. Because wastewater and other analysis tells us that even intelligence-led targeting is capturing only 20-25% of the total volume of drugs entering the country. Tobacco is a similar story. At less than 1% physical coverage, most threats are passing through untouched.

At the same time, we need to rethink where screening happens. Advanced AI-enabled imaging already exists and can scan cargo as it moves through the supply chain — including at the border, inbound and outbound. The question is whether our regulatory model can adapt, to allow industry to operate that capability commercially, with Government being the intelligent recipient of secure imagery and data, rather than the operator of a scanner at a gate. That shift — from border-as-checkpoint to border-as-analytics-function — would transform both coverage and accuracy. Not visibility instead of inspection. But inspection embedded in the supply chain, with AI helping to determine where human judgment is needed.

The system I’m describing needs to do both: see more, and screen more. And it needs to keep working when conditions change — when new threats emerge, when supply chains shift, when the next crisis hits. That’s resilience. And without it, everything else is fragile.

Part 3: Emerging Threats and Legacy Systems

The challenge is compounded by the convergence of threat vectors that was unimaginable when our current systems were designed.

The Integrated Cargo System – the ICS – turned 20 recently. It still processes over one billion transactions annually. It remains critical infrastructure. But it was designed for a world of linear supply chains, paper documentation, and discrete risk categories. Today’s threats don’t fit those categories.

State actors are using criminal proxies to circumvent sanctions and acquire sensitive technologies. Grey-zone operations fall between traditional law enforcement and national security mandates. Cyber-enabled fraud exploits digital trade documentation. Synthetic identities defeat traditional verification systems. The head of MI5 spoke publicly about state actors using private operatives and criminals to do their dirty work. ASIO has documented foreign intelligence services targeting Australia’s defence capabilities and seeking to acquire sensitive technologies through our trade and commercial systems — evading export controls, cultivating insiders, and exploiting the seams between agencies.

A container arrives at an Australian port. ABF holds ICS data. Biosecurity examines cargo declaration data. The AFP might have intelligence on the freight forwarder. AUSTRAC might have financial intelligence on the importer. ASIO might have national security concerns about an associated entity. ACIC exists to fuse criminal intelligence across agencies — but even with that mandate, no single system provides a unified, real-time picture at the border.

The integration that does occur happens through established mechanisms — for example the ACIC Board, Joint Organised Crime Taskforces, and counter-terrorism arrangements. That architecture matters and it should be the foundation we build on. But it relies too heavily on informal relationships, manual information sharing, and goodwill across agency boundaries. It works at the level of specific operations and relationships. It doesn’t scale.

At times our national coordination arrangements can be more art than science — too dependent on who knows whom, and not sufficiently formal, accountable, or fast enough for contemporary threats. When Ironside revealed the extent of border infiltration, it wasn’t because any single system had flagged it. It took a covert international operation to expose what was hiding in plain sight across multiple systems.

The Simplified Trade System — announced as an ambitious whole-of-border reform — has seen its scope narrow from transformation to facilitation. That’s a pattern we cannot afford to repeat.

The lesson is clear: legacy architecture was not designed to see across seams. Each agency maintains its own systems, its own data, its own risk models. This worked when threats were discrete and volumes manageable. It cannot work when threats are convergent and volumes are growing exponentially.

Part 4: Technology, AI, and National Resilience

So what does a modern border technology architecture look like? I want to propose four principles – and then share something I’ve been developing that I think could change how we frame the entire conversation.

Federated, Not Centralised

The solution is not one giant system that does everything. That approach is expensive, brittle, and politically difficult. Instead, we need a federated architecture – a system-of-systems that connects existing platforms, enables secure data sharing, and preserves the distributed responsibilities that reflect our constitutional and administrative arrangements. The key insight is that integration doesn’t require centralisation. What it requires is common standards, secure interfaces, and governance frameworks that determine who can access what data, under what conditions, for what purposes.

Visibility as the Decisive Capability

In high-volume, high-ambiguity environments, visibility is decisive. The ability to see risk before it materialises at the border determines whether interventions can be timely, targeted, and effective. You cannot inspect your way to security when you’re examining less than one percent of containers.

Operation Tin Can demonstrated what’s possible. We brought together 43 countries and some key industry partners, who provided real-time supply chain data. The result: detection and seizure of nearly 100 tonnes of cocaine in six weeks. That was more than an operation – it was a proof of concept. It showed what government and industry can achieve when we lawfully combine data, intelligence, and knowledge in near real time. It’s time to build on such foundations.

AI as Augmentation, Not Replacement

AI will transform border management. Machine learning models are already improving targeting accuracy. Natural language processing is extracting intelligence from unstructured data. Computer vision is enhancing screening. But AI should augment human judgment, not replace it. The decisions that matter at the border – whether to detain, examine, or refer – remain human decisions. AI’s role is to surface the signals humans might miss, manage volumes humans cannot process, and accelerate the cycle from detection to action.

The guardrails matter as much as the capabilities. Explainability – officers and oversight bodies need to understand why a system flagged something. Accountability – clear lines of responsibility when AI-assisted decisions go wrong. Continuous validation – testing models against evolving threats and emergent biases. And human-in-the-loop for consequential decisions. An AI system that delivers efficiency gains but erodes public trust is not a success.

I’m also conscious that we’re in an arms race. Criminal networks and hostile states are using AI too – to generate synthetic identities, to detect patterns in our enforcement, to find gaps in our systems. We cannot afford to be slower adopters than our adversaries. But speed cannot come at the expense of accountability.

Trust as Infrastructure

The architecture I’m describing depends on trust at multiple levels: trust between agencies that data shared will be used appropriately; trust between government and industry that collaboration won’t create competitive disadvantage; trust between nations that shared intelligence won’t be compromised; and trust from the public that their data and rights will be protected. Building this trust is not a communications exercise. It’s an engineering challenge — privacy-preserving technologies, access controls, and oversight mechanisms must be designed in from the start.

Part 5: The Value Equation

Before I get to recommendations, I want to share something I’ve been developing since leaving the Commissioner role. I’m going to ask you to engage with it as a work in progress — because I think it addresses a problem we haven’t solved, and I’m genuinely interested in whether it resonates with this room.

Here’s the problem. We have a measurement crisis at the border.

We count seizures. We don’t count what we miss. We measure activity — containers examined, passengers processed, detections made — but we have no agreed way to measure the gap between what we’re catching and what’s getting through. We impose real costs on industry through compliance and delay, but we can’t be clear-eyed about the benefits we’re generating in return. And when a minister or a secretary asks “is the border working?” — nobody can give a rigorous answer. Because we’ve never properly agreed on what the border is actually for.

At the HTX conference in Singapore last year, I presented an algorithm for crisis leadership — a way of anchoring a complex argument in a simple framework. It was a bit of fun, but the audience found it useful. It gave me the idea to try something similar for border management.

I want to suggest the border is for two things: security and prosperity. And that those two things can be measured as Assurance — how confident are we that threats are being identified and managed — and Throughput — how efficiently legitimate goods and people are moving.

The traditional assumption is that these compete. More security means more friction. I think technology breaks that trade-off — but only if we’re honest about the maths.

Both outcomes are driven by shared underlying capabilities. Visibility — how much of the flow can you actually see, how deep is the data, how timely is it. Precision — what’s your strike rate, how well are you stratifying risk from noise. And Resilience — how adaptable is the system, what redundancy exists, how well does it learn. Response and Capacity then drive Assurance and Throughput respectively — but all of them multiply through Visibility and Precision first.

That multiplicative relationship is the key insight. A ten percent improvement in Visibility doesn’t just add ten percent to assurance — it flows through every downstream function. And the corollary is equally important: when Visibility and Precision are low, no amount of physical inspection can compensate. Wastewater and other analysis indicate we’re capturing only 20-25% of the total volume of drugs entering the country. At less than 1% physical coverage, even intelligence-led targeting is leaving most threats untouched. The framework explains exactly why — and points clearly to where investment will deliver the greatest return.

Required physical coverage should be a function of Visibility and Precision. The better you can see and the more accurately you can target, the less you need to physically touch. That’s not a theory. It’s a mathematical relationship.

And here’s where it becomes a policy argument, not just an analytical one.

Under current settings, the value created by investments in visibility accrues to the system as a whole — but cannot be captured by those who make the investment. A port operator who invests in real-time container tracking, or a logistics provider who builds higher-integrity supply chain data, creates value for everyone but captures almost none of it. That’s a market failure. And it means we’re leaving private capital on the table that could be doing public work.

If we translate assurance into what I’m calling “assured trade” pathways — where demonstrably low-risk, high-visibility flows earn materially faster clearance, lower inspection rates, and priority processing — the business case for private investment becomes compelling and the value becomes capturable. And where industry takes on screening functions within their own supply chains — scanning containers, verifying documentation, maintaining chain-of-custody data — that feeds directly into system-wide Visibility and builds Resilience, because the system is no longer solely dependent on government capacity.

This is where legacy Trusted Trader concepts — the WCO’s AEO framework, the SAFE Framework of Standards — can genuinely evolve. Instead of periodic accreditation, assurance becomes continuously evidenced through data, behaviour, and network context. The EU’s digital product passport requirements begin rolling out from next year — structured, verifiable information moving with goods across borders. The data environment is changing whether we drive it or not. The question is whether our border systems are designed to use it.

And the argument becomes more potent, not less, as the world becomes more complex. In a multipolar environment — where sanctions lists can change with little warning, where the same supply chain can be compliant in one jurisdiction and suspect in another, where trade itself has become a tool of statecraft — a border system that can only measure what it catches is flying blind. Assurance and Throughput, continuously measured and evidenced, give government and industry a common language for navigating that complexity in near real time. The framework scales with the threat.

The investment case and the national security case align. That’s rare in public policy. We shouldn’t waste it.

This framework is nascent — as the slide says, a work in progress. But I think it gives us something we’ve been missing: a way to measure border effectiveness that connects strategic intent to operational reality, and that gives industry a genuine reason to invest alongside government rather than simply absorb its costs.

I’m keen to hear whether that logic resonates — particularly with the port operators, logistics providers, and technology companies in this room.

Part 6: Strategic Recommendations

Let me close with four recommendations — drawn from my Occasional Paper — and I want to be direct: everyone in this room has a stake in whether these happen.

One: Develop a National Border Strategy

Australia needs a strategy that unifies efforts across economic security, national security, and national resilience. Not owned by a single portfolio, but embedded at the highest levels of government. The Aviation White Paper’s recognition of aviation as critical infrastructure signals a shift in how we understand nationally significant systems. The border should be next — and it needs a strategy to match.

Two: Establish an Expert Advisory Board

Border modernisation requires expertise that doesn’t reside entirely in government. An Advisory Board comprising government, industry, and academic representatives should guide strategy development, AI adoption, regulatory innovation, and horizon scanning. Industry holds critical knowledge about supply chains, technology capabilities, and operational realities. We need structured mechanisms to bring that knowledge into policy development — not as lobbying, but as genuine co-design. This room is exactly the talent pool that board should draw from.

Three: Design a National Border Visibility Platform

This is the centrepiece. A federated digital backbone that enables shared situational awareness, coordinated decision-making, and AI-supported risk assessment across portfolios. It should build on existing foundations, incrementally replace the Integrated Cargo System, and create secure compartments for trusted industry collaboration. Think of it as a common operating picture for the border — not owned by any single agency, but accessible to all who need it, with appropriate access controls and audit trails. The technology exists. The challenge is governance, funding, and sustained political will.

Four: Establish a National Border Investment Framework

Border technology investments are currently made portfolio by portfolio, budget cycle by budget cycle. There’s no system-level mechanism to align priorities, sequence investments, or ensure interoperability. That has to change. We should be willing to reinvest part of the value the border generates — duties, taxes, and the avoided cost of crime — into the digital backbone that protects and enables it. A National Border Investment Framework should drive prioritisation based on public value and national security outcomes, under oversight from a senior cross-government committee. And for industry — if we get the investment framework and the assured trade incentives right, the business case writes itself.

Closing

I began with a personal reflection. Let me end with one.

In 44 years – from walking beats as a London bobby, through the IRA bombing campaigns, through anti-corruption commands, through MH17, through COVID, and six years steering Australia’s border – I’ve never seen a more challenging strategic environment. The threats are more complex, more convergent, and more persistent than at any time in my career. The volumes are growing, the adversaries are adapting, and the margins for error are narrowing.

But I’ve also never seen more opportunity.

The technologies being developed in this room and by organisations represented here have the potential to shift border management from reactive enforcement to anticipatory risk management. To build visibility where we currently have blind spots. To forge partnerships that multiply our capabilities. And – if we get the policy settings right – to create a genuine market for border assurance, where industry investment and public value reinforce each other.

At the Sydney Olympics in 2000, Australia’s border was world-leading. The Integrated Cargo System was about to launch as a breakthrough single-window platform. We led the world. Twenty-six years later, we have an opportunity – and an obligation – to lead again. Not by replicating what we did then, but by building what we need now: a border that is digital by design, federated in architecture, AI-enabled in operation, and trusted by the public it serves.

Everyone in this room has a role to play. The technology vendors developing the next generation of capabilities. The government officials shaping policy and procurement. The operators who will use these systems daily. The border I’m describing won’t build itself. It requires people who understand both the strategic imperative and the technical detail to push, cajole, and persist through the inevitable setbacks.

The question is whether we’ll seize that window, or whether we’ll let it close while we manage the urgent at the expense of the important.

I know which choice I’d make. I hope you’ll make the same one.

Thank you. I look forward to the discussion.