Australia’s Defence Crossroads: From Platforms to Architecture in a Contested Indo-Pacific
As I return from my latest visit to Australia, I am thinking back on my time there and how one might look at the evolving situation for Australian strategic policy. I go to Australia twice a year in my role as a Reearch Fellow at the Sir Richard Williams Foundation and have the challenge, the privilege and the opportunity to write the report highlighting the bi-annual seminars held by the Foundation to highlight ADF developments.
Australia enters the late 2020s facing the most demanding strategic environment since the Second World War. The 2026 National Defence Strategy states this bluntly: Australia will face levels of exposure to force projection and military coercion not seen since 1945. Warning time has collapsed. The Indo-Pacific is now a theatre of persistent coercion and episodic conflict rather than a rules-based backwater, and the comforting fiction that Australia can buy its way out of trouble with long-dated platform programs has been exposed as dangerous self-deception.
At the same time, Canberra has begun to articulate a more serious answer: a national strategy of denial anchored in a fight-tonight posture, a reprioritised northern and near-region focus, and a determination to build sovereign, allied-enabled capability that contributes meaningfully to deterrence rather than simply decorating communiqués. The 2026 NDS and its accompanying Integrated Investment Program commit approximately $425 billion in capability investment over the decade to 2035–36, a $155 billion increase over the 2020 Defence Strategic Update baseline. These are not planning aspirations; they are published funding profiles with year-by-year appropriation figures attached.
Yet there is a gulf between aspiration and reality. The central question is whether Australia can move fast enough, conceptually, organisationally, industrially, to turn promising frameworks into real operational capabilities in this decade, not the 2040s. That question runs through three interlocking themes that have become increasingly clear in recent Australian debates: the shift from platforms to architecture, the emergence of denial, resilience, and industrial mobilisation as core national-defence tasks, and the struggle of a middle power to turn autonomy and alliance into genuine architectural influence rather than managed dependence.
From Platforms to Architecture
For much of the post-Cold War era, Australian defence debates have revolved around platforms. Frigates versus submarines, 4.5-generation fighters versus stealth, tanks versus Army-in-the-North. The implicit assumption was that if Canberra chose the right large platforms, at sufficient cost and in sufficient numbers, the rest of the system would fall into place. That era is over.
The decisive shift is architectural. The central question is no longer what any single ship, aircraft, or vehicle can do. It is how effectively it functions as a node in an integrated web of sensors, shooters, and decision-makers distributed across domains and across allied forces. The vocabulary is now familiar, kill chains, kill webs, mesh fleets, stacked effects, but the underlying reality can still be obscured by buzzwords.
The contrast between the linear kill chain and the kill web is the right starting point. In the classic kill chain, a single platform or tightly bounded unit executes the full cycle of detect, decide, and deliver. In a contested Indo-Pacific battlespace, saturated with long-range sensors, precision weapons, cyber and space effects, that linearity is a vulnerability. Take out the key node and the chain breaks. In a kill web, by contrast, those functions are distributed. Sensing is provided by a heterogeneous mix of satellites, crewed aircraft, uncrewed systems, and fixed infrastructure. Decision support flows through a network of human commanders and AI-enabled C2 nodes. Effects are delivered by whichever shooter, land, sea, air, or undersea, is best placed and authorised at that moment.
Australia’s emerging mesh fleet at sea and its evolving airpower construct both illustrate this shift. The Royal Australian Navy’s Maritime Autonomous Systems Unit (MASU) is an organisational expression of kill-web logic. It treats Ghost Shark, Speartooth, and Bluebottle not as isolated curiosities, but as part of a distributed maritime sensor-and-effects layer designed to extend the reach of scarce crewed platforms, populate key maritime approaches with persistent presence, and generate targeting opportunities at acceptable political and operational risk. The platform in this construct is not the hull alone; it is the combination of autonomous vehicles, communications nodes, data flows, and operators that together form a theatre-wide operational unit.
In the air domain, a similar logic is at work. The F-35A is less significant as a fighter in the classical sense than as a survivable sensor and C2 node inside an allied kill web; the E-7 Wedgetail is valuable not as an airborne radar truck but as an integrator and director of effects across a joint formation; Growler and the Next Generation Jammer are disruptive precisely because they can attack an adversary’s targeting architecture rather than simply individual emitters. The emerging role of collaborative combat aircraft, such as Ghost Bat, is most meaningful when understood as added layers in a stacked, multilayered airpower construct, not as cheap substitutes for manned platforms. Air Force Chief AM Stephen Chappell has assessed the Ghost Bat as a world-leading capability with the potential to transform Australia from a tier-one small air force into a tier-one medium-sized air force by adding mass and capability without exhausting its highly trained human crews.
This architectural perspective has two immediate consequences. First, it elevates the importance of institutions that can integrate, experiment, and adapt, MASU at sea, airpower enterprises that own both platforms and doctrine, joint targeting organisations able to exploit the full potential of long-range guided weapons. Second, it exposes the limits of procurement cultures built around platform projects rather than system-of-systems evolution. A country can have the right platforms and still the wrong force, if it fails to invest in the connective tissue, C2, data, doctrine, training, and organisational authority, that turns platforms into a coherent architecture.
Airpower practitioner John Conway, whose career spans Cold War Germany to the contemporary strategic environment, identifies three structural problems with how airpower is currently understood: oversimplification, over-emphasis on change, and under-investment in air combat capability. The kill web has been reduced in much public discourse to a communications problem which misses the operational point entirely. Conway’s framework of stacking effects vertical layers of capability that become progressively harder for an adversary to unravel offers the right analytical template for Australian airpower development, encompassing continuous sensing improvement, the targeting enterprise, long-range strike from sovereign bases, and the affordability frameworks that measure value in outcomes rather than input costs.
Denial, Resilience, and the Fight-Tonight Posture
The 2026 National Defence Strategy represents a significant, if still incomplete, attempt to align Australia’s force design with a realistic appraisal of its strategic circumstances. At its core is a strategy of denial: the deliberate effort to deny any adversary the ability to project military power against Australia through its northern and maritime approaches, and to hold at risk the forces and infrastructure that would enable coercion or attack.
This marks a break from the expeditionary, contribution-based mindset that dominated during the stabilisation campaigns in the Middle East and Afghanistan. Then, force design and investment priorities were shaped by the need to contribute niche capabilities to coalition operations far from home, often from sanctuary bases and with ample warning.
Today, the demand signal is different. Australia must assume that crises in the Indo-Pacific will develop quickly, that its bases and infrastructure will be early targets, and that its ability to rely on unimpeded access to U.S. enabling capabilities may be constrained in time, space, or political will. Professor Justin Bronk of the Royal United Services Institute delivered a stark assessment in a presentation to the Sir Richard Wilklias Foundation seminar audience: Australia faces a two-to-five-year threat timeline, not the previously assumed five to ten years.
Three elements of the emerging posture are worth highlighting. First, geography and range. Denial at distance requires genuine long-range strike and ISR capabilities that can reach into the approaches and chokepoints through which hostile forces would move, and can do so persistently. The 2026 NDS and IIP commit $28–35 billion to targeting and long-range strike, pursuing a doctrinal reorientation that is remarkable in scale: sea-based strike capability is to increase from 120 kilometres to 2,500 kilometres; land-based strike from 40 kilometres to 1,000 kilometres. This is not incremental improvement. It is the operational embodiment of the strategy of denial, explicitly holding potential adversary forces at risk at distances that matter strategically.
Second, resilience and survivability. Bases, ports, fuel, logistic nodes, communications networks, and space assets can no longer be treated as benign background assumptions. They are targets. A fight-tonight posture demands that northern bases be hardened, dispersal options created, fuel and munitions stocks deepened, and theatre logistics—including medical support and repair capacity, treated as war-winning determinants. The $13–16 billion Northern Bases investment, hardening and expanding RAAF Bases Darwin, Learmonth, and Tindal, reflects a force design that assumes its bases will be contested rather than sanctuaries. The Cold War DNA of survive-to-operate practices—dispersal, cross-servicing, routine training to operate under attack—can be recovered, but only if built into exercises, doctrine, and daily routines, not left to occasional experiments.
Third, industrial mobilisation and fighting depth. Denial strategies that rely heavily on long-range missiles, complex munitions, and attritable uncrewed systems are only as credible as the industrial and supply chains that sustain them. Matt Jones of BAE Systems crystallises the historical lesson: waiting for crisis to justify investment leaves Australia with money but no time. The contrast with Ukraine is instructive: Ukraine’s advantage was not its starting industrial capacity but that it began mobilisation eight years before the 2022 invasion, cultivating a culture of innovating at wartime speed. Australia’s Guided Weapons and Explosive Ordnance enterprise, funded at $26–36 billion, addresses the munitions depth problem directly but the gap between rhetoric and capacity remains large, and hard decisions about stockpiling, surge capacity, and willingness to pay for inventories that may never be used in anger lie ahead.
The broader point is whether Australia can generate real fighting depth, the ability not just to fight the first fortnight of a high-intensity conflict, but to sustain operations, adapt, and replenish in the face of losses and disruption. AVM (Retd) Robert Denney states the requirement with clarity: the force must be prepared not only to fight tonight, but to fight tomorrow night, next week, and next month. That requires both domestic industrial capacity and a network of reliable partners prepared to prioritise Australia in war, not simply in peacetime supply arrangements.
Autonomy, Intelligent Mass, and the Economy of the Exchange Ratio
In this context, autonomy and uncrewed systems are often presented as a technological panacea: a way for middle powers to compensate for small inventories of exquisite platforms by fielding swarms of cheap drones and uncrewed vehicles. As with most panaceas, this is misleading. Autonomy can be transformative, but only if framed operationally rather than fetishised technologically.
For Australia, autonomy’s real promise lies in creating intelligent mass: distributed, networked presence that is persistent, politically scalable, and operationally reversible. MASU’s creation institutionalises what has until now been a collection of experiments and projects scattered across the fleet. This is not a reorganisation of boxes on an org chart. It is a strategic decision about how Australia intends to fight at sea in the coming decade. In every domain where autonomous systems have moved from novelty to operational reality, the decisive turning point has been the moment a standing unit took ownership because ownership means resources, career pathways, and doctrine that survives contact with peacetime bureaucracy.
Ghost Shark, as an extra-large autonomous underwater vehicle, addresses a structural problem facing every middle power with limited submarine numbers: how to generate continuous undersea presence across vast distances when the crewed submarine fleet is always oversubscribed. Its modular payload bays allow persistent ISR in areas where crewed submarines are unavailable or politically constrained, or the deployment of undersea effectors under tight command authority. Bluebottle USVs provide persistence at negligible cost, months on station, extending the ISR picture, providing cueing for crewed and uncrewed strike assets, and acting as communications relays between deeply submerged systems and distant command nodes. Speartooth provides an intermediate undersea option, a modular testbed for trialling new payloads and operating concepts before committing to integration at scale.
Together, these missions leverage what autonomy is genuinely good at: persistence, attritability, operating in dull, dirty, or dangerous environments, and generating data at scale. They also align with the economy of the exchange ratio, the requirement that the cost of defence per salvo, per engagement, or per day of presence not exceed the cost imposed on the attacker. Defence analyst Marcus Hellyer makes the point with characteristic directness, deploying the analogy of the bolt-action rifle: if you look at maritime autonomous systems as prototypes of some future perfected capability, you miss the utility they offer right now in their own context. Autonomous systems today are not prototypes of some future fleet. They are capable systems that can begin delivering operational effects against threats that are already evident.
The asymmetric cost logic runs through the lessons of recent conflicts. Ukraine faces thousands of incoming drones monthly alongside hundreds of cruise and ballistic missiles. The cost-exchange ratio of using multi-million dollar missiles to intercept drones costing a few thousand dollars is a losing proposition against adversaries employing saturation tactics. The Advanced Precision Kill Weapon System, laser guidance for standard rockets at a fraction of the cost of dedicated interceptors, illustrates the right response logic. For Australia, the corollary is that replacing exquisite, scarce platforms with exquisite, scarce autonomous systems defeats the purpose. The value lies in tipping the cost curve: forcing an adversary to expend expensive weapons or scarce ISR assets to neutralise cheap, distributed nodes or, ideally, to live with the uncertainty that they cannot neutralise them at all.
But here again, the architectural point is decisive. Autonomy becomes transformational only when embedded in a coherent organisational construct, units that own both the systems and the doctrine, control centres that can fuse data and retask assets dynamically, and legal and policy frameworks that define acceptable levels of autonomous decision-making, especially for systems operating at range with offensive payloads. For a middle power like Australia, the autonomy imperative is therefore less about fielding swarms than about building the institutions and architectures that turn a manageable number of uncrewed systems into usable operational mass.
Sovereign, Allied-Enabled Capability and Middle-Power Agency
Australia’s strategic bet remains clear: its security is inseparable from that of the United States and like-minded partners, yet its interests are not always identical and its geography imposes distinct requirements. The challenge is to reconcile sovereignty and alliance in a manner that enhances, rather than undermines, both.
The concept of sovereign, allied-enabled capability offers a way through. In the air domain, this means building a force in which key elements, F-35, E-7, P-8, Growler, Ghost Bat, long-range strike, are deeply interoperable with US and allied systems, but are maintained, sustained, and increasingly upgraded through domestic or ally-diversified industrial capacity.
Sovereignty in this sense does not imply autarky or rejection of foreign technology. It implies the ability to employ, sustain, and, where necessary, adapt critical capabilities at national discretion, including when allies are heavily engaged elsewhere or politically constrained. AM Chappell frames the deterrence logic precisely: credible deterrence requires capability that is credible, comprehended, and communicated collectively and consistently. Adversaries are deterred not by plans on paper but by the visible, regular, and cumulative demonstration of credible hard power.
At sea, MASU’s integration into AUKUS Pillar II discussions is an opportunity to turn operational experience into architectural influence. If Australia can demonstrate real expertise in long-range undersea autonomous operations in demanding waters, it earns a voice in shaping the standards, data architectures, and command-and-control doctrines that will govern allied autonomous maritime operations. The contracts signed in April 2026 for the first three Mogami-class frigates built by Mitsubishi Heavy Industries with subsequent vessels to be constructed at the Henderson Defence Precinct in Western Australia reflect the same tension: speed of delivery against sovereign industrial base development. Canberra wants ships in the water by the late 2020s and wants to develop the domestic capacity to build them continuously thereafter.
More broadly, Australia is part of a larger story about middle powers in an era of contested orders. Traditional middle-power theory. predicated on multilateral institutions, rules-based order, and niche contribution to US-led coalitions, is strained by the return of major power rivalry. The real constraint on middle-power agency is increasingly the architecture of dependence: who controls critical technology stacks, undersea cables, data flows, critical minerals processing, semiconductor supply chains, and dual-use industrial capacity.
Australia’s efforts to diversify supply chains, develop critical-minerals strategies, participate in trusted semiconductor ecosystems, and build a more resilient defence-industrial base are therefore not economic policy at the margins; they are strategic moves aimed at loosening constraining architectures and building enabling ones. The same is true of its push for Indo-Pacific security webs, overlapping networks of cooperation with Japan, India, Southeast Asian states, Pacific partners, and European navies, that complement, rather than replace, the US-centred hub-and-spoke system. Japan is itself undergoing a parallel strategic pivot, investing heavily in counter-strike capabilities and reshaping alliance architecture; Australia’s deepening bilateral relationship with Tokyo, formalised in the Mogami memorandum, is one concrete expression of a regional security web that goes beyond the traditional bilateral alliance frame.
The key point is that meaningful agency comes from contribution. States that bring genuinely valuable capabilities, concepts, or industrial capacity to allied tables can shape outcomes; those that do not are shaped by them. Australia’s pursuit of sovereign, allied-enabled capability is in that sense not a hedge against the alliance, but a way of making the alliance more robust and better suited to the Indo-Pacific. Whole-of-society defence is the necessary complement: Colonel David Beaumont identifies four pillars, industry, workforce, social cohesion, and institutional decision-making capacity, each of which must be deliberately cultivated in peacetime to ensure the nation has the strategic endurance to weather a protracted conflict.
The Race Between Transformation and Vulnerability
All of this points to a stark conclusion: Australia is in a race between military transformation and growing strategic vulnerability. The direction of travel in policy documents and major investment decisions is broadly sound. Canberra recognises the need to move from platforms to architecture, from expeditionary contributions to denial and resilience, from rhetoric about autonomy and industry to real operational mass and mobilisation potential, and from passive alliance consumption to architectural influence.
But time is not on Australia’s side. The 2026 NDS states openly that at the start of 2024, more states were engaged in armed conflict than at any point since 1946, and that China’s military advances, nuclear-capable air-launched ballistic missiles, hypersonic anti-ship missiles, stealth attack drones, AI-enabled systems, are moving at pace. The risk is not that Australia will choose the wrong broad strategy, but that it will move too slowly, allowing the window in which credible denial can be established to close before the transformation is complete.
The choice set then narrows to unattractive extremes: either over-reliance on allies in conditions where their own freedom of action is constrained, or acceptance of growing vulnerability to coercion. ACM (Retd) Mark Binskin frames the structural challenge: the global rules-based order that has underpinned international stability is under unprecedented strain, stripping away the luxury of extended warning times. As LTGEN Sue Coyle notes, the synchronisation of effects across the space, cyber, and electromagnetic domains will likely be the deciding factor in who prevails—not platforms alone.
Avoiding that outcome demands a more ruthless focus on implementation. That means prioritising investments that strengthen architecture over those that simply add more platforms; treating northern basing, fuel, munitions, and theatre logistics as urgent national projects rather than long-term infrastructure plans; accelerating the operationalisation of autonomy through MASU-like institutions in other domains and through realistic, demanding exercises; making hard choices in procurement to avoid spreading limited resources across too many lines of effort that cannot all be brought to maturity in time; and deepening industrial partnerships that offer real surge capacity and co-development, not just licensed assembly or peripheral workshare.
The 2026 NDS and IIP represent the most ambitious defence planning exercise Australia has undertaken in the post-Cold War era. The scale of investment is real. The conceptual framework, denial, distributed operations, intelligent mass, sovereign capability, is right. What is not yet certain is whether the institutional machinery, the acquisition culture, the industrial base, and the willingness to accept near-term costs for long-term security can be brought to bear at the speed the strategic environment demands.
Australia has begun to ask the right questions about its defence posture. The task now is to answer them with the necessary speed, discipline, and seriousness. The architecture of its future force and its ability to shape, rather than be shaped by, the Indo-Pacific security environment will be determined less by what it intends than by what it can make real before the next crisis arrives. The choice, ultimately, is between adaptation and vulnerability.
Note: These conclusions are based both on the last two Sir Richard Williams Foundation seminars, in September 2025 and April 2026 and my interviews conducted while visiting Australia to support the seminars.