What do model reports say about their ChemBio benchmark evaluations? Comparing recent releases to the STREAM framework

TL;DR

This paper analyzes recent AI model reports on ChemBio safety evaluations, comparing them to the STREAM framework, and highlights areas for improved transparency and best practices in reporting methodologies.

Contribution

It provides a comparative analysis of top AI model reports against the STREAM standard, identifying gaps and recommending improvements for transparency in ChemBio evaluation reporting.

Findings

All reports included some useful details not present in others

Reports generally lacked detailed examples of test materials

There is a need for more comprehensive reporting of elicitation conditions

Abstract

Most frontier AI developers publicly document their safety evaluations of new AI models in model reports, including testing for chemical and biological (ChemBio) misuse risks. This practice provides a window into the methodology of these evaluations, helping to build public trust in AI systems, and enabling third party review in the still-emerging science of AI evaluation. But what aspects of evaluation methodology do developers currently include -- or omit -- in their reports? This paper examines three frontier AI model reports published in spring 2025 with among the most detailed documentation: OpenAI's o3, Anthropic's Claude 4, and Google DeepMind's Gemini 2.5 Pro. We compare these using the STREAM (v1) standard for reporting ChemBio benchmark evaluations. Each model report included some useful details that the others did not, and all model reports were found to have areas for development, suggesting that developers could benefit from adopting one another's best reporting practices. We identified several items where reporting was less well-developed across all model reports, such as providing examples of test material, and including a detailed list of elicitation conditions. Overall, we recommend that AI developers continue to strengthen the emerging science of evaluation by working towards greater transparency in areas where reporting currently remains limited.