The U.S. Army reportedly blew through a year of AI tokens in roughly one month
Annual_Judge_7272 · reddit · 2026-07-22
The U.S. Army reportedly burned a year of AI tokens in about a month
A WIRED report says the U.S. Army announced unlimited tokens for generative AI in May 2026, but the centralized pool was fully exhausted by mid-June, forcing usage caps back on.
The subscription reportedly included 100 million tokens per year, with some users initially getting 200,000+ tokens per month plus automatic top-ups. The Army Enterprise LLM Workspace, powered by Ask Sage, gives access to models such as ChatGPT, Gemini and Llama.
The broader takeaway is that token economics has become the real bottleneck. As usage scales to thousands or millions of users, “unlimited AI” quickly turns into a budget problem, especially for agentic workflows that burn through huge context windows.
Related event: US military exhausts annual AI token budget in a month(2 posts)→
More from Infra
- PyTorch details 2.12x faster on-device streaming ASR on Arm SME2 via ExecuTorch — PyTorch · 2026-10-27
- Revenue per GPU can rise as token prices fall, argues Gavin Baker — GavinSBaker · 2026-10-27
- Anthropic S-1: only $19B total compute spend since inception, overbuild warning — bookwormengr · 2026-10-27
- CoreWeave, You.com and NVIDIA lift BrowseComp 22.9% in 3-week RL post-training sprint — RichardSocher · 2026-10-27
- Grid queues make power architecture an AI-factory decision; Exowatt and Redwood offer fixes — ingliguori · 2026-10-27
- Two CVSS 9.9 Wasmtime flaws allow sandbox escape, one ARM64-only since 2020 — basedjensen · 2026-10-27