A new Security Nexus Perspective by Dr. Elizabeth Vaughan Moyer examines the strategic application of large language models in Indo-Pacific security operations while addressing technology competition and sovereignty concerns.

Moyer, a DKI APCSS fellow and major in the U.S. Air Force, tackles the urgent challenge facing security professionals: leveraging artificial intelligence for enhanced decision-making without compromising critical thinking or accountability. The analysis emphasizes a key argument: organizations that embrace experimentation with imperfect technology will gain significant advantages over those waiting for ideal solutions.

The article introduces a practical “start human, end human” framework, which preserves human judgment at the beginning and end of decision-making processes while using large language models as cognitive tools for information processing. Moyer explores real-world applications, such as maritime domain awareness in the Philippines, and addresses implementation challenges, including language barriers, bias, hallucinations, and sovereignty concerns amid U.S.-China technology competition.

Moyer concludes that successful adoption of large language models requires deliberate sequencing, starting with individual experimentation, progressing to organizational integration, and expanding to regional cooperation. She emphasizes that while technologies follow predictable cycles of hype and disillusionment, hesitation during these transitions carries mounting opportunity costs that Indo-Pacific security organizations cannot afford.

The full linked article was researched and written by the author. This post includes only a summary  with the assistance of artificial intelligence, which were reviewed and edited by DoD personnel to ensure appropriateness and compliance with DoD policies and guidance.