By Jerry Champion
It’s after 1800, and you’re knee-deep in your sixth meeting of the day.
The chat thread in Microsoft Teams is growing longer by the second, sidebar conversations are spiraling, First Alert messages keep spamming your inbox, and you’re somehow expected to incorporate all this late-breaking information into tomorrow’s 0800 Command Update Brief (CUB) and still get home at a decent hour.
Clearly dinner with the family tonight is not going to happen —again!
You get ready to crack open another energy drink and reach for your third Cliff Bar of the day, then you remember you’re not alone. Your AI comrade is right there with you.
So, instead of drowning in the digital chaos, you quietly download the transcript from the recorded Teams session, gather up all those First Alerts, feed it to your AI teammate, and watch as it produces a coherent EXSUM, highlights all assigned tasks, and even creates a properly formatted briefing slide—Times New Roman, font 14, for goodness’ sake. You spot-check your AI-curated products (just like you would any junior staff officer’s work), make some happy-to-glad edits, and then look down at your watch—it’s only 1815.
You walk out the door, making it home in time for both dinner and the kiddos’ bedtime routine. Job well done, Soldier. Welcome to the life of a cognitively enhanced staff officer executing mission command with machines in the age of AI.
While most conversations around artificial intelligence (AI) in the military tend to spotlight advancements like automated ISR target acquisition, algorithmic warfare, decision dominance, or accelerating kill chains, there’s a quieter, deeply practical revolution happening in staff sections across the force. Large language models (LLMs) like ChatGPT (or its DoD accredited cousins CamoGPT, Vantage’s AIP Threads, or Microsoft Co-Pilot) are becoming digital teammates for many weary staff officers drowning in the piles of data heaved upon them in the age of the Fourth Industrial Revolution. These tools aren’t replacing the proverbial Iron Majors who run the Army—they’re augmenting them, providing a reprieve from some of the most mentally exhausting tasks: generating timely and cogent responses to the commander’s relentless request for information, conducting deep research distilled into EXSUMs and white papers, and of course—slides, slides, slides. Why are there still so many slides?
The Meeting Trap and the Power of Asynchronous AI
As staff officers, we’ve all been there: two meetings scheduled at the same time, both labeled “must attend.” One’s a working group that always runs long, the other a dry run with unfinished slides. Just as you’re weighing which meeting to attend, a Soldier knocks on your door—eyes down, voice low and asks, “Do you have a minute?”
Now what?
First and foremost, we fail in our most important duty if we don’t drop everything to care for that Soldier. No staff meeting—no matter how urgent—outranks the well-being of the people we lead.
Traditionally, that’s the dilemma for our Iron Majors: stepping away means missing key decisions, falling behind on product development, and getting blindsided at tomorrow’s sync. It’s the emotional tax of doing the right thing. However, for the AI-Empowered Iron Major here’s where your digital teammate can step into the trenches with you: enable recording and transcription on both meetings, then hand the digital content over to your AI teammate for rapid consumption, analysis, and feedback.
If one of those meetings ended up being a substantive session, you can play it back later at 1.5x speed to ensure you get the message first hand and in full context. If both ended up being low-yielding time thieves, let your AI teammate summarize them for you. This is where LLMs shine. Afterward, feed the transcripts to your LLM teammate and ask for EXSUMs, task trackers, or even draft emails for your stakeholders.
Bonus points—you no longer have to ask your NCOIC to sit in a room full of officers while good idea fairies ruminate their way into next week’s tasking order. You’re welcome, Sergeants Major!
Example 1: Prompt for Meeting Summary
“Summarize this meeting transcript into a 3-paragraph EXSUM. Include decisions made, key themes, and any suspense items or assigned tasks.”
Optional: “Output in Army SITREP format.”
Example 2: Prompt for Task Tracker
“From this meeting transcript, extract all due-outs and action items. Include assigned individuals, deadlines (if mentioned), and format in a table.”
From Raw Notes to Polished Products to Strategic Engagements
Writing is a core competency for any staff officer. But the volume and pace can be relentless. LLMs help bridge that gap. Start with a few bullets, a voice memo, or a hasty brain-dump, and these tools can turn it into a concise point paper, a professional email, or the beloved Memorandum for Record.
Beyond writing from scratch, LLMs are also exceptional at digesting and summarizing source material or scouring the internet for rapidly evolving real-world issues.
Example 3: Prompt to Draft Media Talking Points
“Read all available news reports on the recent earthquake in Hawaii and provide media talking points for the USARPAC commander. Highlight specific areas where people are talking about gaps in the disaster response and provide options where military resources could support the effort. Ensure the output is easily readable in bullet format and fits on a 5×8 index card.”
MDMP with a Digital Staff Collaborator
Planning remains a human-centered process. But large language models can dramatically accelerate the staff work leading up to the first session of the Military Decision Making Process (MDMP). Ask the model to draft an initial problem statement, outline a mission analysis brief, or reformat your ideas into slides or written orders. Prior to official wargaming, these tools can also conduct shaping analysis across common planning frameworks that examine diplomatic, informational, military, economic, financial, intelligence, and law enforcement factors.
Once these factors are considered, ask your assistant to develop tailored evaluation criteria to support the comparison phase—whether quantitative or qualitative.
Working in a classified or closed network environment? That’s ok. Start your work in the unclassified space with unfettered access to open-source material, then transfer it over to the high-side and layer in your classified data to refine the product—it’s not like you’ve never done swivel chair operations before. Even better, your AI teammate just got his security clearance— as these LLM tools are increasingly available on classified networks.
Example 4: Prompt for COA Generation
“Given the recent announcement on the Army Transformation Initiative, and the forthcoming establishment of a Western Hemisphere Command, generate three different COAs for the force structure of this organization.”
Example 5: Prompt for iterative MDMP engagement
“Based on the COAs generated in the previous response, update the three COAs to take into consideration that Western Hemisphere Command will also assume the roles and responsibilities historically assigned to ARNORTH and ARSOUTH. Ensure statutory obligations under Title 10 of the U.S. Code are taken into consideration.”
Working With the Machine: It’s a Conversation, Not a Crutch
No one expects the plan to be complete and orders published after the first MDMP session, so don’t expect these AI tools to produce a finished product from a single prompt. Most complex staff work requires nuance, clarification, and iteration. The same type of process must take place with your AI teammate; we can go faster, but we can’t take shortcuts.
We must treat our AI teammate like a junior staff officer, NCO, or a trusted co-worker. That first prompt is just the beginning of the conversation. The deeper you go—providing context, asking clarifying questions, nudging the response in the right direction—the more refined and valuable your output becomes. Ask follow-up questions, reframe the task, provide examples, and course-correct.
Just like you’d coach a human teammate, you need to guide the machine to excellence. What’s better is this digital teammate does not get tired or annoyed with your incessant questioning, nor does it suffer from the infuriating burden of trying to find white space on each of your calendars.
To be clear, this is not about outsourcing human judgment to machines or replacing critical thinking skills. When used well, these AI tools can serve as Cognitive Exoskeletons, enhancing our natural abilities to achieve previously unimaginable speed, efficiency, and clarity. Used poorly, they become a Cognitive Couch—comfortable, passive, and slowly eroding the mental discipline we rely on as staff officers. Then it’s just a digital sloth disguised as productivity—and is the same kind of indiscipline that would lead your ranger buddy to fall asleep in the patrol base and put the whole team and mission at risk.
Cautions, Context, and Commander’s Intent
Like any tool, LLMs come with usage guidelines:
Security: Never input classified or sensitive data into commercial tools unless they’re DoD accredited.
Judgment: AI outputs can be confident and flat wrong. Always validate before publication.
Ownership: The product still has your name on it. Treat AI as a collaborator, not a scapegoat.
Adapt or Die in the Age of AI
AI isn’t replacing staff officers—yet. But with massive force structure changes like the Army Transformation Initiative reducing manpower across formations—particularly among headquarters staff—we’re being asked to do more with less. The question becomes: how do we make staff officers more effective without burning them out?
Part of the answer lies in redefining what teamwork means and retooling the modern-day staff officer. The one who learns to effectively partner with AI in the mundane trenches of staff warfare—who knows when and how to delegate to their digital teammate—will outperform, outthink, and outlast those who don’t. AI tools like LLMs can provide a reprieve from the grind of scribing meeting minutes, producing slides, and formatting briefings—freeing up time for critical thinking, face-to-face leader engagement, and the rarest of luxuries: time with family.
Since the inception of the Prussian General Staff under Helmuth von Moltke, the institution of the staff has evolved through each technological shift—from railroads and telegraphs to tanks and radios. The emergence of AI is simply our generation’s turn to modernize staff work once again, adapting how we operate to match the speed and complexity of warfare in the Fourth Industrial Revolution.
So when the transcript’s been summarized, the EXSUM submitted, and your CUB slides are complete—and you look down at your watch and realize you’re done before 1900—that’s not a fluke. That’s the future of staff work. You walk out the door with your cognitive ruck a little lighter, your head clearer, the commander better informed, and your kid asking if you want to play Legos before bed.
That’s not science fiction. That’s disciplined initiative—with a digital teammate!
Got a great prompt or AI workflow that saves you hours a week? Share it with your peers. The AI teammate only works if we coach it, collaborate with it, and share it with our fellow suffering servants.
Jerry W. Champion is a Lieutenant Colonel in the United States Army with over 17 years serving as an Infantry and Signal Corps officer. His previous operational assignments include: 82nd Airborne, 75th Ranger Regiment, 3rd Infantry Division, XVIII Airborne Corps, 5th Special Forces Group, National Geospatial-Intelligence Agency, and the White House. He is currently assigned to Headquarters, Department of the Army G-3/5/7 and is the incoming G6 for the 11th Airborne Division—America’s Arctic Angels—in Anchorage, Alaska.
