mlsanityData to 7 Oct 2026

Code World Model Preparedness Report

arXiv 2605.00932 ↗First posted 1 May 2026cs.SE cs.AIcited by 0

On the map: World models → World models for language agents

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in World models for language agents, one of 11 directions in World models. Builds on 15 of its 26 references that are in the map. First posted on arXiv 1 May 2026.

Builds on

15 of 26 references.

References · 15

  1. 2025Qwen3 Technical Reportcited by 8,195
  2. 2024SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineeringcited by 1,628
  3. 2016Concrete Problems in AI Safetycited by 1,582
  4. 2025gpt-oss-120b & gpt-oss-20b Model Cardcited by 1,258
  5. 2024The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearningcited by 488
  6. 2025Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscationcited by 338
  7. 2024Deliberative Alignment: Reasoning Enables Safer Language Modelscited by 292
  8. 2023AI Deception: A Survey of Examples, Risks, and Potential Solutionscited by 185
  9. 2023PaperQA: Retrieval-Augmented Generative Agent for Scientific Researchcited by 143
  10. 2024LAB-Bench: Measuring Capabilities of Language Models for Biology Researchcited by 138
  11. 2024Planning In Natural Language Improves LLM Search For Code Generationcited by 79
  12. 2025Stress Testing Deliberative Alignment for Anti-Scheming Trainingcited by 71
3 more
  1. 2025The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systemscited by 57
  2. 2025Virology Capabilities Test (VCT): A Multimodal Virology Q&A Benchmarkcited by 27
  3. 2024Badllama 3: removing safety finetuning from Llama 3 in minutescited by 10

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025CWM: An Open-Weights LLM for Research on Code Generation with World Models
  2. 2024World Models: The Safety Perspective
  3. 2026Debugging code world models
  4. 2024Generating Code World Models with Large Language Models Guided by Monte Carlo Tree Search
  5. 2026Software World Models: From Consequence Prediction to Decision Value
  6. 2024Software Vulnerability and Functionality Assessment using LLMs
  7. 2026Learning Reasoning World Models for Parallel Code
  8. 2026When a Verified World Model Still Loses: Play-Adequacy vs Prediction-Accuracy in LLM-Synthesized Code World Models