AOI: Turning Failed Trajectories into Training Signals for Autonomous Cloud Diagnosis

Researchers present AOI (Autonomous Operations Intelligence), a multi-agent AI framework that automates Site Reliability Engineering tasks while maintaining security constraints. The system achieved 66.3% success rate on benchmark tests, outperforming previous methods by 24.4 points, and can learn from failed operations to improve future performance.

• →AOI framework enables secure automated operations by using locally deployed models to avoid exposing proprietary data.
• →The system separates read-write operations to prevent unauthorized changes while allowing safe learning from operational trajectories.
• →A 14B parameter locally deployed model outperformed Claude Sonnet 4.5 on diagnostic tasks with unseen fault types.
• →The framework converts failed operations into training signals, improving success rates by 4.8 points and reducing variance by 35%.
• →AOI addresses key enterprise AI deployment challenges including data privacy, execution safety, and continuous improvement from failures.