The DoD killer app with GenAI is….training!

What LLMs do best is process existing documentation to produce new text within very complex interpretative instructions. Where that may be an ancillary part of many job workflows in the DoD, the single best fitting and widest impacting task that function represents is training content production. Every service hires very expensive highly experienced SMEs to support the creation, refinement, and publication of training content for every career field in the military. It is by far the most time consuming and labor intensive task for an expensive and limited talent pool. Training commands could leverage LLMs to both speed content creation as well as improve training content quality across all career fields.

The big advantages here are none of this requires future AI agents that still need to be build, it fits within and leverages the existing human workflows of training content production today, and it’s very low risk as there’s no direct operational interaction with weapon systems or any decision making. Oh and for bonus points the way you build great content for one career field is the same process that can be used for them all, so this can be a HUGE capability enhancement in a short period of time.

You simply will not find a lower risk, easier integration, and higher impact return on a single use case for GenAI anywhere in the DOD today.

Plenty of hopes and dreams in AETC for GenAI with a whole host of existing personnel ready and willing, but this still remains a massive win waiting to happen. They just need systems and approvals…that’s a very solvable problem.