Agentic Experience (AX) as a Force Multiplier in Human-Machine Teaming
Technical Concept Paper
1. ABSTRACT
The evolution of human-machine interaction has moved from User Experience (UX), focused on reactive, user-driven interfaces to Agentic Experience (AX), where autonomous AI agents act proactively, collaboratively, and dynamically based on high-level human intent.
In Special Operations Forces (SOF) contexts, AX offers the potential to:
- – Reduce cognitive load on operators;
- – Compress the OODA loop (Observe–Orient–Decide–Act);
- – Enhance decision superiority in uncertain, time-critical, and resource-constrained environments
- – Integrate multi-source ISR data to generate dynamic courses of action (COAs).
Challenges include maintaining meaningful human control, ensuring trust and explainability, addressing cyber vulnerabilities, and updating doctrine and training. Early adoption of AX represents a potential asymmetric force multiplier for SOF.
2. Background and Rationale
2.1 From UX to AX
Traditional UX places the human operator as the primary executor, with systems providing reactive support. While effective in stable environments, UX has limitations in dynamic, high-pressure operations.
AX introduces:
High-level intent reasoning
- Automated task decomposition and planning
- Tool utilization and workflow orchestration
- Multi-agent collaboration for decision support
2.2 Relevance to Special Operations Forces
- SOF operate in small teams under degraded communications and high uncertainty
- Cognitive overload can threaten mission success
- AX can act as a force multiplier, enhancing operational tempo and effectiveness
3. Conceptual Framework
| Feature | UX | AX | Operational Impact |
| Task Execution | Human-driven | Agentic, autonomous | Reduces operator workload |
| Decision Support | Reactive | Proactive, predictive | Accelerates OODA loop |
| Data Handling | Manual | Multi-source fusion | Improves situational awareness |
| Adaptation | Static | Context-aware, dynamic | Enables real-time adjustments |
Key Principle: Operators define “what”, agents manage “how”.
4. Potential Applications in SOF
Mission Planning
- Commander sets high-level intent (e.g., night infiltration, extraction in 48 hours)
- AX agents integrate multi-source ISR data, simulate scenarios, generate fragmentary orders (FRAGOs)
- Planning time reduced from days to minutes
Execution and Real-Time Adaptation
- Real-time monitoring of sensor feeds and anomaly detection
- Autonomous adjustment of routes, drone deployment, countermeasures
- Operators remain focused on tactical decisions
Intelligence and Decision Support
- Multi-source data fusion and threat prioritization
- Dynamic SITREPs and COA recommendations
- Supports rapid, context-aware decision-making
5. Benefits and Challenges
Benefits:
- Increased operational speed and tempo
- Scalable for small, distributed teams
- Resilient to degraded connectivity
- Reduced cognitive load on operators
Challenges: – Maintaining meaningful human control in lethal actions – Explainability and operator trust in AI systems – Cybersecurity and potential agent “hallucinations” – Need for doctrinal evolution and training focused on intent orchestration.
6. Recommendations and Next Steps
- Conduct empirical trials in simulated and controlled operational environments
- Develop trust and explainability metrics for agentic AI systems
- Establish ethical and doctrinal frameworks for AX integration
- Explore modular, low-SWaP-C architectures compatible with existing SOF platforms
7. Conclusion
AX represents a paradigm shift in human-machine teaming. For SOF, agentic systems have the potential to act as a force multiplier, enabling small teams to generate disproportionate operational effects. Early experimentation and adoption may determine decision superiority in 21st-century battlefields.
8. Disclaimer
This concept paper reflects a conceptual analysis based on publicly available sources and does not represent official doctrine or actual operational capabilities.
References
Xi, Z., et al. (2023). The rise and potential of large language model based agents: A survey. arXiv preprint arXiv:2309.07864.
Brown, A., Kandasamy, J., Van Roo, B., & Evans, R. (2026). Agentic AI and the Pentagon’s integration challenge. War on the Rocks.
Friar, J. M. (2025). Agentic AI: Strategic adoption in the DoD (CSIAC Report).
Norman, D. A. (2013). The design of everyday things (Revised and expanded edition). Basic Books.
Osinga, F. (2007). Science, strategy and war: The strategic theory of John Boyd. Routledge.
Russell, S., & Norvig, P. (2021). Artificial intelligence: A modern approach (4th ed.). Pearson.
(2025). USSOCOM RFI TE 26-2: Agentic Artificial Intelligence (AI)-based modular capabilities for SOF (Special Notice). SAM.gov.
Wang, L., et al. (2024). A survey on large language model based autonomous agents. Frontiers of Computer Science.
Wooldridge, M. (2023). The road to conscious machines: The story of AI. Pelican Books.