Building the Future of National Security Excellence | 2025-2028
by Uladzimir Laryoshyn <uladzimir.laryoshyn@caci.com>
Current defense contractor roles face varying levels of AI-driven transformation
Existing defense contractor roles experiencing transformation through 2028
Impact: Augmentation not replacement. AI handles 30% of design validation through Model-Based Systems Engineering. Lockheed's AI Factory enables 8,000+ engineers to leverage AI for end-to-end development.
Skills Needed: Digital engineering proficiency, AI/ML model integration, data architecture expertise, MBSE tools, digital twin technology.
Impact: 20-30% productivity gains through AI-assisted coding. 1.79M national workforce. Developers supervise AI code generation rather than writing from scratch. Booz Allen reports permanent productivity increases.
Skills Needed: Prompt engineering for code generation, AI model integration, MLOps, security for AI systems. Air-gapped coding assistants for SIPR/JWICS.
Impact: Starkest disruption. AI automates 55-65% of routine threat detection. Some organizations replace entire 80-person teams with AI. Entry-level faces high displacement; senior roles remain in demand. 189,300 positions nationally, 33% projected growth through 2034.
Market Driver: CMMC 2.0 (Dec 2024) requires 350,000 DIB firms to demonstrate cybersecurity maturity. Salary bifurcation emerging.
Impact: AI automates 50-60% of initial data processing and pattern recognition. Pentagon's $200M contracts with Google, xAI, Anthropic, OpenAI target intelligence workflows. Project Maven processes imagery intelligence autonomously.
Role Evolution: Transition from data collection toward strategic interpretation, understanding adversary intent, identifying patterns across disparate sources.
Impact: AI automates 60-70% of routine coordination tasks. 67% of supply chain executives report full/partial AI automation in 2025. DoD's Replicator Initiative prioritizes autonomous logistics.
Role Evolution: Coordinators become AI system supervisors and exception handlers. Strategic sourcing specialists remain in medium demand; transactional roles face significant pressure.
Impact: Highest displacement risk across all defense roles. 60-80% of test execution now automatable. Manual testing declined 70-80% where AI deployed. Raytheon's flight test automation dramatically reduces manual cycles.
Survival Strategies: Specialize in test automation architecture, transition to DevSecOps, focus on hardware/physical testing, move into AI model validation.
Impact: AI handles 25-50% of research, data analysis, and documentation. Booz Allen's AI consulting track trains non-technical consultants for AI-enabled data storytelling. Role bifurcates: AI-enabled consultants thrive, AI-resistant struggle.
Market Context: DoD eliminates 60,000 civilian positions through 2026, contractors capture work previously performed in-house. Shift to outcome-based contracting favors AI productivity.
Impact: Low displacement risk because role focuses on designing automation systems—precisely the skillset needed in AI-enabled environments. Demand increases as organizations deploy more AI systems requiring orchestration.
Impact: AI-powered network optimization automates routine monitoring and configuration. Core architecture expertise remains valuable; routine operations face automation.
Impact: 40-50% of documentation automatable through AI. Technical writing and compliance documentation increasingly AI-generated with human oversight.
Impact: Hands-on physical work remains irreplaceable. Role transforms through AI augmentation: predictive analytics for maintenance scheduling, XR-guided repairs, digital twin integration for diagnostics.
Distinct AI-focused roles currently in active recruitment
Purpose: Emerged in response to DoD's 2022 Responsible AI Strategy. Ensure AI systems align with DoD's five principles: Responsible, Equitable, Traceable, Reliable, Governable. Unlike commercial AI ethics, must address international humanitarian law, rules of engagement, and autonomous weapons ethics.
Skills: Technical AI/ML understanding, DoD AI Ethical Principles, bias detection/mitigation, international humanitarian law, algorithmic auditing.
Clearance: Secret to TS/SCI depending on programs.
Purpose: Design, test, and refine prompts for LLMs deployed in classified environments. Commercial LLMs refuse military queries; defense needs custom LLMs (Scale AI's Defense Llama) requiring specialized prompting. DoD's Task Force Lima deploys generative AI for mission planning and intelligence synthesis.
Skills: LLM architecture understanding, military terminology/doctrine, prompt optimization, classified system deployment.
National Baseline: $62K-$95K; TS/SCI cleared positions command premium.
Purpose: Generate artificial datasets replicating real-world defense scenarios without exposing classified information. Critical bottleneck: AI training data availability in classified environments. Generate synthetic imagery for autonomous weapons, simulated sensor data, photorealistic environments for computer vision.
Impact: Reduces AI development costs up to 80% compared to real-world data collection.
Skills: Generative AI (GANs, diffusion models), computer graphics, domain expertise (radar, EO/IR sensors), data privacy/security.
Purpose: Conduct adversarial testing on defense AI systems to identify vulnerabilities before deployment. China and Russia actively develop capabilities to attack U.S. military AI. Develop attack scenarios specific to military contexts (autonomous vehicle spoofing, threat detection evasion).
Skills: Adversarial machine learning, penetration testing, AI security, red teaming, defense system knowledge.
Clearance Premium: AI expertise + security knowledge + clearance commands top compensation.
Purpose: Design workflows and interfaces enabling effective collaboration between military personnel and AI agents. DoD emphasizes augmentation rather than replacement. Develop trust-building protocols and training programs for human-AI collaboration.
Focus Areas: Fighter pilot-AI wingman teaming (Loyal Wingman), intelligence analyst-AI assistant workflows, commander-AI decision support.
Skills: Human factors engineering, AI system design, military operations, cognitive psychology, UX for high-stakes environments.
Purpose: Fine-tune large language models for defense-specific applications on classified networks. Adapt commercial LLMs to military domains, train models for military planning queries commercial models refuse. Highly specialized due to GPU-constrained classified networks and limited cleared ML engineer pool.
Skills: LLM architecture (transformers), distributed training, model compression/optimization, SIPR/JWICS deployment, military domain knowledge.
Purpose: Create virtual replicas of defense systems, weapons platforms, and operational environments. 73% of A&D organizations established long-term digital twin roadmaps. Implement real-time data integration, enable predictive maintenance, simulate mission scenarios.
Market Context: McKinsey estimates linking physical/digital worlds could generate $11.1T annually. A&D digital twin investment up 40% YoY.
Skills: IoT/sensor integration, physics-based modeling, real-time data streaming, 3D visualization, AI/ML for predictive analytics.
Purpose: Ensure defense contractor AI systems comply with DoD requirements, federal regulations, and acquisition standards. DoD's 2023 Data, Analytics, and AI Adoption Strategy requires comprehensive governance. Contractors must demonstrate Responsible AI compliance for contract awards.
Skills: DoD acquisition regulations, AI policy frameworks, risk management, CDAO requirements, audit/compliance.
Purpose: Distinct from traditional cybersecurity. Implement security measures protecting defense AI models from adversarial attacks. Develop defensive distillation techniques, design secure model deployment pipelines. Traditional cybersecurity focuses on networks; AI security addresses AI-specific attack vectors.
Skills: Adversarial ML, model robustness, secure enclaves, cryptography for ML, threat modeling for AI systems.
Purpose: Manage end-to-end integration of AI capabilities into defense systems and workflows as AI moves from proof-of-concept to operational deployment. Coordinate across engineering, operations, and acquisition teams.
Skills: Program management, AI/ML technical knowledge, DoD acquisition, change management, systems integration.
Salary Range: $120K-$180K mid-level, $150K-$250K+ senior AI program managers.
Roles potentially necessary as AI systems mature and scale
Purpose: Translate DoD Ethical AI Principles and Directive 3000.09 into executable technical specifications for autonomous weapon systems. As autonomous weapons move from concept to deployment (Loyal Wingman, autonomous maritime vehicles, AI missile defense), need specialists bridging policy, ethics, and technical implementation.
Skills: International humanitarian law, rules of engagement, technical AI/ML architecture, safety-critical systems engineering, policy interpretation, scenario modeling for edge cases.
Purpose: Manage "teams" of AI agents working collaboratively on defense missions. Once individual AI systems prove valuable, operations will deploy multiple specialized AIs requiring coordination. Coordinate data processing, target recognition, mission planning, logistics optimization AIs working together.
Skills: AI system orchestration, multi-agent systems, game theory, military operational planning, API integration, conflict resolution between AI recommendations.
Purpose: Critical immediately as AI systems need data across classified/unclassified network boundaries. AI requires massive data access; defense data exists across multiple security classifications. Traditional Cross-Domain Solutions handle file transfers; AI needs real-time data streaming, model synchronization, result sharing.
Skills: Cross-domain solution architecture, AI/ML pipeline engineering, security clearance/accreditation processes, network architecture across SIPR/NIPR/coalition networks.
Purpose: Investigate AI system failures, accidents, and anomalous behaviors in defense applications. Unlike human error, AI failures involve complex interactions between training data, model architecture, environmental conditions, adversarial actions. First major AI incident will create immediate demand.
Skills: Digital forensics, reverse engineering, AI/ML model interpretability, military accident investigation protocols, adversarial ML attack detection.
Purpose: Optimize cognitive load distribution between humans and AI systems in operational environments. As soldiers manage multiple drones plus autonomous ground vehicles, or analysts supervise multiple AI systems, cognitive load becomes critical operational constraint.
Skills: Human factors engineering, cognitive psychology, AI system behavior modeling, military operations analysis, workload assessment methodologies.
Purpose: Predict and assess adversary AI capabilities as China and Russia deploy operational AI systems. Combines intelligence analysis with technical AI expertise and technology forecasting—rare combination. Focus on second and third-order effects of Chinese and Russian AI programs.
Skills: Intelligence analysis, OSINT, technical AI/ML expertise to assess capability claims, understanding of Chinese/Russian defense industry, technology forecasting.
Purpose: Manage human workforce impacts from AI deployment as displacement reaches significant scale. Identify displaced roles, design retraining programs, create new career pathways, ensure equitable AI adoption. Currently largely unplanned despite inevitable restructuring.
Skills: Change management, organizational development, adult learning theory, training design, labor economics, workforce analytics, AI capability assessment.
Purpose: Develop hybrid systems as quantum computing reaches "practical utility" for specific defense applications. Pentagon FY2026 allocates $2.2B for AI + quantum with explicit focus on convergence. Quantum offers advantages for optimization problems, enhanced sensing, cryptography.
Skills: Quantum computing principles/programming (Qiskit, Cirq), classical AI/ML expertise, optimization algorithms, understanding of NISQ devices, quantum-safe AI security.
Purpose: Design systems building appropriate trust levels between operators and AI systems—addressing both over-reliance and under-utilization. Critical for operational effectiveness as AI systems scale.
Purpose: Maintain, debug, and update older AI models no longer fully understood but remaining operational in critical defense infrastructure. As AI systems age, understanding their behavior becomes challenging.
Purpose: Design AI-enabled operations across air, land, sea, space, and cyber domains simultaneously for JADC2 (Joint All-Domain Command and Control). Requires understanding of multi-domain warfare and AI orchestration.
Purpose: Resolve conflicts when multiple AI systems provide contradictory recommendations in time-critical situations. Design decision frameworks for AI conflict resolution.
Purpose: Develop brain-inspired computing architectures offering advantages in power efficiency for edge deployment in autonomous systems and sensors.
Purpose: Manage high-stakes scenarios where AI systems malfunction during operations. Provide "break glass" authority and emergency decision-making protocols for AI failures in combat.
Clearance premiums offset but don't eliminate compensation gap
79% of cleared workers job-hunting RIGHT NOW. We're not competing—we're bleeding out.
"It's NOT the pay—we pay 12% MORE than tech. It's everything else."
At 13% turnover, we lose $312M+ annually in recruiting, training, clearance processing, and lost productivity.
CACI's 90-Day Talent Blitz
Target: Cut turnover from 13% to 6% by end 2025. Savings: $156M annually + competitive advantage worth 10x that.
Non-Negotiable Truth: Palantir pays LESS than CACI for AI roles but has 90%+ retention. They win on culture, mission, and treating engineers like engineers—not PowerPoint operators. We have every advantage except execution. FIX IT.
Internal development of core technical roles
Aggressive external recruitment
DoD SkillBridge program - massive untapped opportunity
Academic collaborations create talent pipelines
Competitors are moving now. Delay equals market share loss.
COMPLIANCE DEADLINES ARE NOT NEGOTIABLE
BUILD THE COMPETITIVE MOAT
SEPARATE FROM THE PACK
Reality Check: Palantir, Anduril, and Booz Allen are executing these strategies NOW. Every quarter of delay costs market position we may never recover.
Lockheed, Northrop, Raytheon, Boeing
Their Advantages:
Their Challenges:
24,000 employees | Intelligence Community Focus
CACI's Key Advantages:
CACI's Strategy:
Palantir, Anduril, Shield AI
Their Advantages:
Their Challenges:
By 2028, industry leaders will have pulled decisively ahead
CEO-level ownership driving transformation across the organization
Purpose as primary motivator, not just compensation - meaningful work over bureaucracy
Give technical talent meaningful technical problems to solve, not PowerPoint duty
Treat workforce transformation as strategic imperative with dedicated resources
Partner with DoD, academia, and industry on shared challenges
CACI's AI workforce transformation will define our competitive position in the $832 billion+ defense market
With 24,000 employees and deep intelligence community expertise, we have the foundation to lead
The time to act is now.