US military mandates deployable AI within 30 days of release, not waiting for perfect models

AI Agent 不必等模型完美：美军给 AI Builder 的部署方法

The US Department of War's 2026 AI memo requires new models to reach deployable status within 30 days of public release, arguing that the delay of waiting for perfect models outweighs the risk of imperfect alignment. The article lays out deployment guardrails: constrain agent action boundaries first (read-only/sandbox), pause for human confirmation at critical decision points, verify behavior with execution receipts rather than self-reports, and make authorization dynamic with fast rollback. The post does not name specific models or performance numbers—the focus is on operational resilience, not model scores.

Why it matters: The DoD's 2026 AI memo mandates 30-day deployability for new models, and the article delivers four concrete guardrail layers rather than vague principles — directly useful for anyone shipping agents. Score held back because no specific model or performance numbers are disclose...