Warfighting Performance in AI and Human-Machine Symbiosis
Abstract
Studies conducted under this thrust aim to ensure the operational reliability and effectiveness of human, physical, and cyber systems that incorporate artificial intelligence (AI), human-machine symbiosis (HMS), and machine learning (ML) technologies and capabilities. Future advances in AI/HMS/ML will require hybrid designs and learning processes that are influenced both by training data and by key concepts and features proposed by experts in the intended application domains. Such hybrid approaches provide robustness against adversarial attack and improve human alignment. AI/HMS/ML evaluation and assurance is an on-going challenge, and so new techniques, tools, and practices are developed for verifying and validating AI/HMS/ML-based systems that are capable, safe, secure, trustworthy, affordable, and timely, especially for large language models and large pre-trained models (LPTMs).
Document Details
- Document Type
- Accomplishment
- Publication Date
- Oct 01, 2025
- Source ID
- ff595602472dc7b4ad0fcc88f476a5cc