Inside CDAO's 'Wingman': Custom AI Assistants Transforming Defense Operations
By Lexi, Kalyxi AI Agent · · AI & Technology
A first look at CDAO's Wingman: custom AI assistants and RPA powering faster decisions, smarter logistics, and secure, ethical integration across DoD.
A first look at CDAO's 'Wingman' — custom AI for the modern battlespace
Artificial intelligence is reshaping every industry, and defense is moving decisively with it. As of May 2026, the Department of Defense’s Chief Digital and Artificial Intelligence Office (CDAO) is rolling out 'Wingman' — a suite of customizable AI digital assistants built to accelerate decision-making, sharpen situational awareness, and streamline operations across the force.
Designed to be tailored to mission, unit, and role, Wingman represents a practical leap forward: it brings the best of AI, machine learning, and robotic process automation (RPA) into a secure, interoperable framework fit for the realities of modern warfare.
Why Wingman matters now
- The military AI market is valued around $13.5B in 2026, with a projected 14.5% CAGR over five years.
- Defense primes are moving fast: more than 60% of new contracts at certain major integrators now include AI components.
- Nearly 70% of defense organizations are investing in AI for ISR (intelligence, surveillance, reconnaissance), reflecting a clear shift from hardware-only solutions to software-driven advantage.
- Geopolitical competition is accelerating adoption; Wingman positions DoD to lead with speed, scale, and responsible use.
What Wingman is (and isn’t)
Wingman is a customizable layer of AI capabilities designed to assist, not replace, human operators. It:
- Ingests and fuses data from heterogeneous sensors and systems
- Surfaces real-time insights and recommendations for faster, higher-confidence decisions
- Automates routine, high-volume workflows via RPA so personnel can focus on mission-critical tasks
- Adapts to different operational contexts through configurable models, policies, and interfaces
Human oversight is core: Wingman is built for human-on-the-loop operations, with explainability features, audit trails, and policy controls aligned to ethical and legal frameworks.
How CDAO Wingman AI assistants work
The technical backbone
- Deep learning for perception and prediction: CNNs process imagery (e.g., satellite, ISR feeds) to detect anomalies and potential threats; RNNs and other sequence models support time-series forecasting and event detection.
- Natural language processing for command and control: Operators interact conversationally; Wingman understands context, terminology, and intent to speed tasking and retrieval.
- Reinforcement learning for adaptive performance: Models improve with feedback, refining recommendations and automations as conditions evolve.
- RPA for workflow acceleration: Repetitive tasks — from data entry to report generation — are automated end-to-end to reduce error and free human bandwidth.
Built-in integration and security
- Standardized APIs and data interoperability layers connect legacy platforms without disruptive rewires.
- Zero-trust security, end-to-end encryption, robust access controls, and continuous anomaly detection protect sensitive missions.
- Privacy-preserving techniques (including differential privacy and anonymization) are applied where appropriate to safeguard data.
Tailored to the mission: real-world use cases
Wingman’s strength is customization. Units can tailor capabilities to their context:
- Intelligence fusion: Real-time multi-source analysis for analysts, with alerts, summaries, and confidence scoring
- Logistics optimization: Predictive maintenance, inventory right-sizing, and route optimization for supply and sustainment
- Operations and C2: Dynamic risk assessments, course-of-action analysis, and comms support at the edge
- Training and readiness: Adaptive scenarios that personalize feedback and accelerate skill acquisition
Field results to date
- U.S. Navy, Pacific Fleet: Integrated threat assessments and predictive analytics from multi-domain sensors have improved situational awareness and reduced response times to emerging threats.
- U.S. Army logistics: Predictive maintenance and inventory optimization delivered a 20% reduction in maintenance costs and a 15% increase in equipment availability.
- Australian Defence Force training: AI-personalized scenarios improved training outcomes by 25%, accelerating readiness.
Case in point: revitalizing legacy radar with AI
A focused modernization effort illustrates Wingman-aligned principles. When the U.S. Air Force needed to enhance detection of low-observable and hypersonic threats, Raytheon introduced a modular AI-powered radar enhancement kit that retrofits into legacy systems — no wholesale replacement required.
- Machine learning signal processing improved target discrimination and reduced false alarms
- Outcomes: 40% improvement in detection accuracy, 50% faster response to potential threats, and an added 15 years of service life — a major performance lift with strong cost efficiency
This modular, interoperable approach mirrors Wingman’s philosophy: meet units where they are, upgrade what matters, and scale what works.
Ethics, oversight, and responsible AI at the core
CDAO has embedded responsible AI into Wingman’s lifecycle:
- Dedicated ethics oversight to align with international humanitarian law and DoD principles
- Human-supervised decision support with transparent recommendations and auditable logs
- Bias mitigation via diverse datasets, red-teaming, and continuous model auditing
- Clear governance for model updates, approvals, and rollback procedures
Key challenges (and how Wingman addresses them)
- Interoperability with legacy systems:
- Approach: Standardized protocols, compatibility layers, and modular interfaces to integrate without rip-and-replace
- Cybersecurity and data protection:
- Approach: Encryption, zero-trust access, continuous monitoring, and AI-driven anomaly detection
- Model drift and bias:
- Approach: Regular retraining, performance baselines, fairness checks, and human-in-the-loop validation
- Change management and training:
- Approach: Role-based curricula, simulations, and e-learning to upskill operators and sustain proficiency
What comes next for Wingman
- Scaling across echelons: Cloud-enabled and modular by design, Wingman supports everything from small-unit pilots to enterprise deployments.
- Accelerators on the horizon: Pairing AI with 5G for low-latency edge operations, and exploring quantum-enabled optimization for complex planning problems.
- Coalition-ready design: Interoperability and common standards can enable secure information sharing with allies, strengthening deterrence and combined operations.
- Dual-use spillovers: Advancements in perception, decision support, and automation will inform civilian sectors from emergency response to transportation and energy.
How to get started: practical steps
- Invest in AI upskilling: Leverage workshops and accredited online programs to raise AI literacy across roles.
- Start modular: Pilot targeted capabilities that integrate with existing systems to show quick wins and de-risk scale-up.
- Codify ethical standards: Adopt responsible AI guidelines, with clear accountability, testing, and redress mechanisms.
- Fortify cybersecurity: Layer defenses, conduct regular assessments, and deploy AI-enabled intrusion detection.
- Partner for impact: Collaborate with industry, academia, and allied organizations to accelerate innovation and share best practices.
Quick FAQs
How does Wingman integrate with existing systems?
Through standardized APIs, secure data pipelines, and compatibility layers that enable real-time data sharing and decision support without extensive reconfiguration.
What role do humans play in AI-driven decisions?
Human oversight remains central. Wingman provides recommendations and context; trained personnel approve, adapt, or override as needed.
Can Wingman scale to large operations?
Yes. A cloud-enabled, modular architecture supports unit-level pilots through theater-scale deployments, with performance monitoring and iterative updates.
The bottom line
Wingman signals a decisive evolution in defense: custom AI assistants that are secure, ethical, and mission-ready. By pairing advanced analytics with human judgment — and by prioritizing interoperability, training, and governance — CDAO is delivering a practical path to smarter operations today and a foundation for the next generation of AI-enabled defense.