Editorial illustration for Army AI Tool Ask Sage Used to Reclassify Personnel Job Descriptions
Army's Ask Sage AI Tool Reclassifies Job Roles
Army AI Tool Ask Sage Used to Reclassify Personnel Job Descriptions
The Army told its own soldiers to slow down on the AI. In May, Army Chief Information Officer leadership announced unlimited tokens for Ask Sage, the generative AI platform troops use to run models like ChatGPT, Gemini, and Llama on Controlled Unclassified Information. By mid-June, the pool was empty. An email obtained by WIRED and sent to members of the Army's Combat Capabilities Development Command spells out the fallout: usage limits are back, and it's not clear whether the token supply gets renewed after October 1.
The timing is awkward. Roughly a month before that email went out, the Department of Defense had publicly touted that nearly half its 3.5 million employees were using AI on the job. Ask Sage isn't a side project.
It's accredited for handling sensitive unclassified data, used by the DOD's Chief Digital and AI Office for acquisitions work, and deployed across the Army for tasks ranging from drafting documents to reclassifying personnel job descriptions. One Army employee, granted anonymity because they weren't authorized to talk to reporters, described just how fast that year's worth of tokens disappeared.
The Army uses Ask Sage, a multimodal generative AI platform where users can run different large language models (LLMs), including Alphabet’s Gemini, Meta’s Llama, and OpenAI’s ChatGPT. “Apparently the whole Army burned through the whole year of tokens for just one service,” says an Army employee who spoke to WIRED anonymously because they were not authorized to speak to the press.
Why this matters
The token shortage is the interesting detail here, not the reclassification task itself. An "unlimited" pool that runs dry within weeks tells us the Army either badly underestimated demand or badly underestimated cost, and neither is reassuring for an agency now steering hundreds of thousands of employees toward daily AI use. For developers and founders selling into government, this is a preview of the actual constraint that matters: procurement can promise unlimited access, but someone still has to pay per token, and that bill arrives faster than budget cycles adjust.
For researchers, the personnel-reclassification use case is worth watching too. Rewriting job descriptions sounds low-stakes until you remember those descriptions determine pay grades, promotion eligibility, and staffing decisions for real people. If Ask Sage is doing that work with limited oversight and now limited tokens, the quality of those outputs deserves more scrutiny than a rate-limiting email suggests it's getting.
The lesson for anyone building or deploying enterprise AI: usage caps aren't just an IT headache, they're a signal about how seriously the deployment was actually planned.
Related Reading
Research & Benchmarks Google's FACTS benchmark shows 70% factuality ceiling across four tests
Research & Benchmarks Databricks finds multi-step agents beat single-turn RAG by 21% to 38% on STaRK
Research & Benchmarks Nvidia's DLSS 4.5 beta adds 6x Multi Frame Generation for RTX 50 GPUs
From Archives Microsoft adds AMD-powered Azure HXv2 for complex chip design workloads
From Archives NVIDIA TAO Agent Skills Accelerate Vision Model Post-Training
Common Questions Answered
What is Ask Sage and which large language models does the Army use on this platform?
Ask Sage is a multimodal generative AI platform that allows Army troops to run different large language models on Controlled Unclassified Information. The platform supports multiple LLMs including Alphabet's Gemini, Meta's Llama, and OpenAI's ChatGPT, giving soldiers access to various AI capabilities for their operational needs.
Why did the Army reinstate usage limits on Ask Sage tokens in mid-June?
The Army had to reinstate usage limits after the entire year's token supply was depleted in just a few weeks following the announcement of unlimited tokens in May. An Army employee revealed that the whole Army burned through the entire year's token allocation for just one service, indicating severe underestimation of either demand or cost.
What does the token shortage reveal about the Army's AI implementation strategy?
The token shortage demonstrates that the Army either badly underestimated demand or underestimated costs for AI usage across its workforce, neither of which is reassuring for an agency directing hundreds of thousands of employees toward daily AI use. This situation highlights a critical procurement constraint: while government agencies can promise unlimited access to AI tools, the actual resource limitations may not support widespread adoption at the scale intended.
How quickly were the unlimited tokens consumed after the Army announced them in May?
The unlimited token pool was completely exhausted by mid-June, meaning the entire year's supply of tokens was burned through in approximately one month. This rapid depletion underscores the massive demand for AI capabilities within the Army's personnel and operations.
Further Reading
- Army launches AI platform for efficiency and innovation - LinkedIn (Army Deputy Chief of Staff, G-6)
- Army launches Army Enterprise LLM Workspace, the revolutionary AI platform that wrote this article - Army.mil
- Army AI Workspace Launches - YouTube (Huachuca Pulse)
