Code World Model Preparedness Report
For AI safety and governance, this report provides a negative result (no new risks) for a code generation model, but is incremental as it only confirms existing risk levels.
Meta assessed Code World Model (CWM) for catastrophic risks and misaligned propensities, finding no additional frontier risks beyond current AI ecosystem, leading to open-weight release.
This report documents the preparedness assessment of Code World Model (CWM), a model for code generation and reasoning about code from Meta. We conducted pre-release testing across domains identified in our Frontier AI Framework as potentially presenting catastrophic risks, and also evaluated the model's misaligned propensities. Our assessment found that CWM does not pose additional frontier risks beyond those present in the current AI ecosystem. We therefore release it as an open-weight model.