Beyond Rivet Joint and Dragon Lady: The USAF’s Multi-Domain ISR Architecture for Pacific Air Dominance
For decades, two aircraft defined American intelligence superiority: the U-2 Dragon Lady, slipping silently through the stratosphere at 70,000 feet, and the RC-135 Rivet Joint, hoovering up enemy signals from the edges of contested airspace. These were — and in many respects still are — marvels of engineering and tradecraft. But the Pacific theater of the 2020s and beyond is a fundamentally different environment than the one that birthed these platforms, and the adversary waiting across that ocean has studied them carefully.
China’s military modernization has produced one of the most sophisticated anti-access/area denial (A2/AD) networks in history. Long-range surface-to-air missiles, advanced fighter aircraft, and electronic warfare systems are specifically designed to threaten exactly the kind of large, slow, high-value ISR platforms that America has long depended upon. The vulnerabilities are no longer theoretical. Flying an RC-135 or a U-2 within range of modern Chinese air defenses is a strategic calculation that grows riskier with each passing year.
The United States Air Force recognizes this reality. What’s emerging in response is not simply a new airplane or a better sensor — it’s an entirely new architecture for intelligence, surveillance, and reconnaissance. This is the story of how the USAF is building a multi-domain ISR ecosystem designed for the most complex, contested, and consequential operational environment on Earth: the Indo-Pacific.
The Strategic Context: Why the Pacific Demands a Revolutionary ISR Approach
China’s A2/AD Challenge and the Threat to Legacy Platforms
The Indo-Pacific is not Europe. The distances are staggering — from Guam to Taiwan is roughly 1,700 miles; from Oahu to the South China Sea is over 5,000. Any ISR architecture must contend with these vast operational distances while simultaneously threading through an adversary air defense network of unprecedented depth and sophistication.
China’s People’s Liberation Army has invested heavily in integrated air defense systems that extend hundreds of miles offshore. The HQ-9 and S-400 family of surface-to-air missiles can engage targets at altitudes and ranges that make traditional ISR orbits extremely dangerous. Add to this China’s growing inventory of carrier-killer ballistic missiles, electronic warfare capabilities, and a rapidly maturing space-based surveillance system, and the picture becomes clear: the days of permissive ISR environments near the first island chain are over.
High-Value, Low-Density — and High-Risk
The USAF’s ISR legacy has been built around “exquisite” platforms — extraordinarily capable systems produced in small numbers at tremendous cost. The RC-135 Rivet Joint fleet numbers just 17 aircraft. The U-2, first flown in 1955, has a total inventory of roughly 27 airframes. These are irreplaceable national assets, and that’s precisely the problem.
When a single platform represents billions of dollars in investment, decades of operational data, and enormous intelligence-gathering capability, losing even one to hostile action is a strategic disaster. An adversary who understands this calculus doesn’t need to shoot down a Rivet Joint — the credible threat of doing so can deny access entirely, achieving the same intelligence-denial effect without firing a single missile.
This “high-value, low-density” vulnerability isn’t merely tactical. It represents a fundamental mismatch between the ISR architecture inherited from the Cold War and the demands of great power competition in the Pacific.
The Speed Imperative
Modern warfare moves at a pace that legacy ISR processes struggle to match. The Pentagon’s concept of “kill chain compression” — reducing the time from sensor detection to effector engagement — demands ISR data measured in minutes, not hours. Traditional workflows, where imagery or signals data is collected by an aircraft, transmitted to a ground station, analyzed by human intelligence professionals, and eventually passed to decision-makers, are simply too slow for the hypersonic missile age.
The USAF needs an ISR architecture that collects, processes, and disseminates intelligence at machine speed, enabling commanders to act inside an adversary’s decision cycle. Building that architecture is the central challenge of American air power in the Pacific.
Core Pillars of the Multi-Domain ISR Architecture
The USAF’s answer to these challenges isn’t a single platform or program — it’s a layered, interconnected system of systems built on four foundational pillars: a distributed sensing grid, advanced data processing and fusion, resilient networking, and sophisticated human-machine teaming.
Pillar One: The Distributed Sensing Grid
The most conceptually significant shift in USAF ISR thinking is the move from a handful of exquisite, centralized sensors to what Air & Space Forces Magazine has termed a “sensing grid” — a vast, distributed network of diverse, dispersed, and in some cases expendable sensors spanning every domain.
The logic is elegant and militarily sound. If your ISR capability depends on a few large platforms, an adversary can defeat it by threatening those platforms. But if your ISR capability is distributed across hundreds or thousands of nodes — some manned, some uncrewed, some space-based, some ground-based — the adversary’s problem becomes exponentially harder. You can’t hold a grid at risk the way you can hold an aircraft at risk.
Next-Generation Manned Platforms
The transition isn’t purely to uncrewed systems. The Boeing E-7A Wedgetail, which the USAF is acquiring to replace a portion of its aging E-3 Sentry AWACS fleet, brings a quantum leap in airborne battle management and surveillance capability. With its Multi-Role Electronically Scanned Array (MESA) radar and advanced data link capabilities, the E-7 can simultaneously track hundreds of air and surface contacts while serving as a node in the broader sensing grid.
The P-8 Poseidon, while primarily a Navy maritime patrol aircraft, integrates critical anti-submarine warfare and maritime surface surveillance data into the joint ISR picture — particularly valuable in a Pacific scenario where sea lines of communication and submarine operations are central to any conflict.
The Uncrewed Revolution: From Workhorses to Attritable Sensors
The RQ-4 Global Hawk has served as the high-altitude, long-endurance backbone of USAF uncrewed ISR for years. But even the Global Hawk — which costs roughly $131 million per aircraft — represents a high-value asset that can be threatened in contested airspace. The future lies in something fundamentally different: attritable platforms.
Attritable ISR means sensors cheap enough that their loss in combat is operationally acceptable. Dozens of these platforms, each carrying capable sensors, can saturate an adversary’s defensive picture and still deliver actionable intelligence even if some are shot down. This philosophy represents a profound cultural shift for an institution that has historically treated each ISR platform as precious.
The MQ-9 Reaper, for all its effectiveness in permissive environments like counterterrorism operations, lacks the stealth and survivability for contested Pacific scenarios. Its operational concept is being rethought accordingly, with greater emphasis on standoff operations and payload flexibility rather than close-in penetration.
Collaborative Combat Aircraft: The Sensor Node Revolution
Perhaps no program better exemplifies the new ISR architecture’s logic than the Collaborative Combat Aircraft (CCA). The USAF has signaled plans to acquire thousands of AI-enabled uncrewed aircraft designed to operate alongside manned fighters. While the CCA concept spans multiple mission areas including strike and electronic warfare, its potential as a distributed sensor node is transformative.
Imagine a package of F-35s and F-22s approaching a contested area, surrounded by a swarm of CCAs — each one acting as an expendable sensor, pushing data back through the network. The CCAs collect signals intelligence, track surface targets, and monitor adversary air activity, feeding that data into the fusion architecture. If a CCA is shot down, the swarm redistributes and continues the mission. The manned aircraft never need to enter the most dangerous threat envelopes.
Space-Based ISR: The High Ground Gets Higher
Space has always been a critical ISR domain, but the shift from a few large, expensive geosynchronous satellites to proliferated low-Earth orbit (LEO) constellations is changing the calculus entirely. Where a single GEO satellite might cost billions and take years to replace, a constellation of smaller, cheaper LEO satellites provides persistent coverage through disaggregation — the same resilience logic that drives the distributed sensing grid at the airborne level.
The National Reconnaissance Office and US Space Force are actively developing new satellite constellations optimized for hyperspectral imaging, radio frequency sensing, and missile warning — all critical capabilities for the Pacific theater. Commercial satellite imagery providers like Planet Labs and Maxar are delivering near-real-time coverage that supplements government systems, providing scale and resilience that no classified program could afford to replicate on its own.
Hyperspectral sensors, which can detect objects and activities invisible to standard optical or radar systems, are particularly valuable for monitoring Chinese naval bases, underground facilities, and missile sites. RF sensing satellites can locate and characterize radar emissions across vast ocean areas, helping to build the electromagnetic order of battle that used to be the exclusive domain of platforms like the Rivet Joint.
Sea and Ground Nodes: Integrating the Full Picture
The sensing grid doesn’t stop at the atmosphere. Naval platforms — from Aegis-equipped destroyers with their powerful SPY-6 radars to attack submarines with their passive sonar arrays — contribute critical data to the shared ISR picture. Ground-based radar networks in Japan, South Korea, and Guam extend the sensing grid’s reach into the first and second island chains.
Cyber and electromagnetic sensors round out the picture, probing adversary networks and monitoring the electromagnetic spectrum for indicators of military activity. This cyber-ISR dimension, while often classified, represents one of the most valuable intelligence streams in any modern conflict scenario.
Pillar Two: Advanced Processing and Fusion — Making Sense of the Data Deluge
Collecting intelligence is only half the challenge. The distributed sensing grid described above will generate data volumes that are simply impossible for human analysts to process using traditional methods. A single Global Hawk mission can generate more imagery than a team of analysts can review in weeks. Scale that to hundreds of sensor nodes across an entire theater, operating continuously, and the data processing challenge becomes existential.
Artificial Intelligence and Machine Learning as Force Multipliers
This is where AI and machine learning transition from buzzwords to battlefield necessities. Project Maven, launched by the Department of Defense in 2017, began as an effort to apply AI to drone footage analysis — specifically, automating the identification of vehicles, people, and activities that human analysts were drowning in. Its evolution represents the leading edge of how AI is being embedded into ISR workflows.
Modern AI applications in ISR span several critical functions:
– Automated Target Recognition (ATR): Machine learning models trained on millions of labeled images can identify aircraft, ships, missile launchers, and other military equipment with speed and consistency that no human analyst can match. What takes an analyst minutes per image takes an AI milliseconds.
– Pattern-of-Life Analysis: AI systems can monitor locations over time, building behavioral baselines that flag anomalies — a ship that normally follows a predictable schedule suddenly deviating, or a airfield where activity patterns change before a known exercise. This predictive intelligence is enormously valuable.
– Sensor Fusion: Perhaps most critically, AI can correlate data from multiple, disparate sensor types — imagery, radar returns, signals intelligence, and cyber indicators — into a coherent operational picture that no single sensor could provide. This multi-source synthesis is where the real intelligence value lives.
Edge Computing: Getting Intelligence to the Fight Faster
Traditional ISR processing involves collecting data, transmitting it to distant ground stations (often in the continental United States), having analysts process it, and then sending finished intelligence back to the theater. This chain introduces latency that modern warfare cannot tolerate.
Edge computing addresses this by pushing processing capability closer to the sensor — onto the aircraft itself, into forward-deployed computing nodes, or into the cloud architectures supporting deployed forces. An AI-equipped sensor node that can identify a target, correlate it with other data streams, and pass actionable intelligence to a shooter within seconds of detection represents a capability leap that fundamentally compresses the kill chain.
Pillar Three: Resilient Networking — The Connective Tissue of Multi-Domain ISR
A distributed sensing grid filled with advanced AI processing is useless if the network connecting it can be disrupted. Adversaries know this, and degrading communications networks — through electronic warfare, cyber attack, and direct action against communication satellites — is a centerpiece of Chinese military doctrine.
JADC2: The Overarching Framework
Joint All-Domain Command and Control (JADC2) is the Pentagon’s overarching concept for connecting sensors to shooters across all military domains — air, land, sea, space, and cyber — in near real time. It’s ambitious to the point of being described by some analysts as the most complex military integration challenge in history, but its logic is sound: in a multi-domain fight against a peer adversary, the side that can share information fastest and most reliably across all domains will have a decisive advantage.
ABMS: The Air Force’s JADC2 Contribution
The Advanced Battle Management System (ABMS) is the USAF’s specific contribution to JADC2. Rather than a single system, ABMS is best understood as an evolving framework — a set of standards, architectures, and technologies designed to enable rapid data sharing and network integration across Air Force and Space Force platforms and systems.
ABMS demonstrations have showcased the ability to use data from space-based sensors to cue aircraft to targets, share that targeting data across services in near real time, and compress what was historically an hours-long process into minutes. In a 2020 demonstration, ABMS connected satellites, ground radars, fighter aircraft, and ship-based weapons systems into a common operational picture — a glimpse of what the fully realized architecture could accomplish.
Multi-Path Communications and Cyber Resilience
The networked ISR architecture must be able to operate when portions of the network are degraded or destroyed. This demands multi-path communications — leveraging diverse links including military satellite constellations, commercial satellite communications, airborne mesh networks, and terrestrial fiber where available — so that the failure of any single link doesn’t collapse the network.
Cybersecurity is inseparable from this networking challenge. ISR data is only valuable if it can be trusted, and an adversary who can inject false data into the ISR network or deny access to it has effectively blinded the commander. Cyber resilience — the ability to continue operating under adversary cyber attack — must be designed into the architecture from the ground up, not bolted on as an afterthought.
Pillar Four: Human-Machine Teaming — The Intelligence Behind the Intelligence
Technology enthusiasts sometimes discuss AI-driven ISR as if it implies the removal of humans from the process. The reality is more nuanced and more interesting: the goal is human-machine teaming, where AI handles the high-volume, repetitive processing tasks while human analysts and commanders focus on higher-order reasoning, contextual judgment, and decision-making.
A human analyst freed from reviewing ten thousand frames of uninteresting imagery can focus on the anomaly the AI flagged, apply deep regional and cultural knowledge to interpret it, and present a commander with a confident assessment rather than a data dump. The human becomes a force multiplier when properly augmented, not an obstacle to automation.
Decision superiority — providing commanders with faster, more accurate, and more actionable intelligence than their adversaries possess — is the ultimate output of the entire ISR architecture. Every sensor node, every AI algorithm, every network link exists to serve this single purpose.
Key Technologies and Programs Driving the Transformation
Several specific programs are shaping the future ISR architecture more than any others:
Collaborative Combat Aircraft (CCA) — With the USAF signaling intent to acquire potentially thousands of these AI-enabled platforms, CCAs will fundamentally reshape the sensor grid. Manufacturers including General Atomics and Anduril are competing to deliver aircraft that can autonomously execute ISR missions, adapt to changing threat environments, and operate as loyal wingmen to manned aircraft.
Project Maven’s Evolution — Moving well beyond its original drone footage application, Project Maven’s AI/ML capabilities are being extended to integrate broader ISR data streams and support theater-level intelligence analysis. The lessons learned from Maven are being embedded into the ISR processing architecture.
Open Architecture Imperatives — The USAF has learned painful lessons from proprietary, closed systems that resist rapid technology insertion. The new ISR architecture is being built on open standards that allow new sensors, algorithms, and processing capabilities to be integrated quickly as technology evolves — critical in a competition where the adversary is continuously improving.
Future Stealth ISR Platforms — While details remain classified, the USAF is known to be pursuing next-generation penetrating ISR concepts that could operate inside denied airspace where even advanced CCAs would be at risk. These stealthy, potentially uncrewed platforms would serve as the “tip of the spear” for the sensing grid, collecting data that no standoff sensor could reach.
The Role of Allies: Extending the Sensing Grid
America doesn’t face the Pacific challenge alone, and the multi-domain ISR architecture explicitly incorporates allied capabilities. Japan’s F-2 and next-generation F-X fighters, Australia’s E-7A Wedgetail (which the USAF is now matching), and South Korea’s advanced surveillance systems all contribute data to the shared intelligence picture.
The challenge is interoperability. Sharing sensitive ISR data with allies in near real time requires common data standards, agreed security frameworks, and the political will to actually share intelligence rather than hoard it. The JADC2 architecture must accommodate coalition operations, and exercises like Talisman Sabre and Pitch Black are increasingly used to test and refine these combined ISR workflows.
The Five Eyes intelligence-sharing alliance — encompassing the United States, United Kingdom, Australia, Canada, and New Zealand — provides a foundation for signals intelligence sharing that predates the current competition but is being actively updated for the demands of the new ISR architecture.
Challenges and Obstacles on the Path Forward
Articulating an architectural vision is considerably easier than building it. The USAF’s multi-domain ISR transformation faces substantial obstacles across technical, budgetary, cultural, and human dimensions.
Technical Hurdles
Data standardization remains a fundamental challenge. Sensors from different manufacturers, services, and nations produce data in different formats. Building the fusion capability that makes the sensing grid work requires either standardizing those formats — a massive coordination effort — or building translation layers capable of integrating disparate data streams. Both approaches carry technical risk.
Bandwidth limitations in the Pacific are a persistent constraint. The vast distances and potential disruption of satellite communications create scenarios where the network architecture must function with severely degraded connectivity. Designing ISR systems that degrade gracefully rather than catastrophically is an unsolved problem.
Budget Realities
Transforming the ISR architecture costs money that competes directly with other USAF priorities — fighters, bombers, space systems, and nuclear modernization. The Pentagon’s defense budgets, while substantial, are not infinite, and the ISR transformation requires sustained investment over years before it delivers full capability.
Congressional support matters enormously. Programs like ABMS have faced skepticism from legislators who see expensive demonstrations without clear operational timelines. Building and maintaining congressional confidence in this multi-year, multi-billion-dollar architectural transformation is itself a strategic requirement.
Cultural Resistance
Perhaps the most underappreciated challenge is cultural. Military institutions are naturally conservative, and the shift from platform-centric to data-centric ISR thinking represents a genuine paradigm change for an institution whose identity has long been tied to specific aircraft. The ISR community’s culture — built around the RC-135 crew that flew through Soviet radar coverage or the U-2 pilot who operated alone at the edge of the atmosphere — must evolve without losing the expertise and dedication those traditions represent.
Training and Workforce Transformation
The multi-domain ISR architecture demands new skill sets at every level. Intelligence analysts must become proficient with AI tools, understanding both their capabilities and their failure modes. Network engineers must understand ISR requirements. CCA operators must develop entirely new concepts of employment for distributed, AI-autonomous sensor nodes.
The USAF is building these capabilities, but training pipelines are slow and the competition for technically skilled personnel is fierce — both with other government agencies and with the private sector. An AI developer who could work for a technology company at twice the government salary requires compelling reasons to serve in uniform.
Ethical Considerations
The integration of AI into ISR and the kill chain raises genuine ethical questions that the military takes seriously, even if public discussions sometimes overlook them. Automated target recognition systems can be wrong, and errors in a lethal context are qualitatively different from errors in a commercial application. Ensuring meaningful human judgment remains in consequential decisions — while still achieving the speed advantages that make the architecture valuable — requires careful policy design that the DoD is actively working through.
Conclusion: Information Advantage as the Foundation of Pacific Deterrence
The U-2 Dragon Lady and RC-135 Rivet Joint will remain in service for years to come, and their contributions to American security cannot be overstated. The Dragon Lady has been gathering intelligence since 1956 — longer than most of the people making decisions about its replacement have been alive. These are extraordinary machines, crewed by extraordinary people.
But the nature of the threat has changed, and the ISR architecture must change with it. The USAF’s multi-domain vision — a distributed sensing grid spanning airborne, space, sea, and cyber domains, integrated through resilient networks and accelerated by artificial intelligence, culminating in human-machine teaming that delivers decision superiority to commanders — represents the most ambitious intelligence architecture ever attempted by any military force.
The strategic logic is compelling. An adversary facing a sensing grid that cannot be held at risk by threatening a handful of large platforms, that processes data faster than humans can respond, and that maintains connectivity despite electronic warfare and cyber attack, faces a fundamentally different challenge than one facing the ISR architecture of twenty years ago.
Getting there requires sustained investment, institutional courage to challenge platform-centric thinking, close coordination with allies, and the patience to let a complex, distributed architecture mature into operational capability. None of that is guaranteed. But the alternative — relying on exquisite, vulnerable platforms against a peer adversary who has studied and prepared for exactly that approach — is not a strategy. It’s a vulnerability.
The future of USAF ISR in the Pacific won’t be defined by any single platform. It will be defined by the architecture that makes every sensor smarter, every network more resilient, and every commander better informed than their adversary. That’s the contest that matters.
—
Frequently Asked Questions
Why are the U-2 Dragon Lady and RC-135 Rivet Joint no longer sufficient for Pacific ISR?
Both platforms represent Cold War-era ISR designs that excel in permissive or semi-permissive environments. In the Pacific, China’s advanced A2/AD network — including long-range surface-to-air missiles, electronic warfare systems, and advanced fighters — poses a credible threat to large, relatively slow, high-altitude aircraft operating in or near Chinese air defense engagement zones. Their “high-value, low-density” nature also means that even the credible threat of loss can deny their operational use, effectively granting the adversary the intelligence-denial effect without firing a shot.
What is the “sensing grid” concept the USAF is building toward?
The sensing grid is a distributed network of diverse sensors spanning multiple domains — airborne platforms (manned and uncrewed), space-based satellites, naval systems, ground-based radars, and cyber/electromagnetic sensors — that collectively provide persistent, resilient ISR coverage. Unlike a few exquisite platforms, the grid is designed so that the loss of any individual node degrades but doesn’t collapse the overall capability. This distribution makes it far harder for an adversary to achieve ISR denial through threats to specific platforms.
How does JADC2 relate to the USAF’s ISR architecture?
Joint All-Domain Command and Control (JADC2) is the overarching DoD framework for connecting sensors to shooters across all military domains in near real time. The USAF’s Advanced Battle Management System (ABMS) is its service-specific contribution to JADC2. For ISR, JADC2 provides the networking architecture that allows data collected by one sensor — whether a satellite, a CCA, or a ship-based radar — to be rapidly shared with analysts, commanders, and weapons systems across the joint force, dramatically compressing the kill chain.
What role will AI play in the new ISR architecture?
AI and machine learning are essential force multipliers in the new architecture. They perform automated target recognition in imagery, conduct pattern-of-life analysis to identify behavioral anomalies, fuse data from multiple disparate sensor types into coherent intelligence products, and process volumes of data that would be impossible for human analysts to handle manually. Project Maven pioneered many of these applications. The goal is human-machine teaming — AI handles high-volume, repetitive processing while human analysts apply contextual judgment to the AI’s outputs, and commanders receive faster, more accurate intelligence.
What are Collaborative Combat Aircraft (CCAs) and how do they contribute to ISR?
Collaborative Combat Aircraft are AI-enabled uncrewed platforms designed to operate alongside manned fighters. While they serve multiple mission roles, their potential as distributed, attritable sensor nodes is transformative for ISR. CCAs can penetrate contested areas to collect signals intelligence and imagery, feeding data back through the sensing grid network. Because they are designed to be produced in large numbers at lower cost than traditional aircraft, their loss in combat is operationally acceptable — removing the vulnerability that makes high-value ISR platforms strategically risky to deploy in contested environments.
How are allies integrated into the USAF’s multi-domain ISR architecture?
Allied nations with advanced ISR capabilities — particularly Japan, Australia, and South Korea in the Pacific — are increasingly integrated into the shared intelligence picture. Australia and the USAF both operate the E-7A Wedgetail, facilitating interoperability. The Five Eyes intelligence-sharing framework provides a foundation for signals intelligence sharing. The primary challenges are data standardization, security frameworks for real-time sharing, and building the political consensus to actually share sensitive intelligence in crisis situations. JADC2’s coalition operation capability is explicitly designed to accommodate allied participation.