AI Companies Should Report Pre- and Post-Mitigation Safety Evaluations
This addresses the problem of inadequate safety oversight for policymakers and regulators in the AI industry, proposing mandatory disclosures to improve transparency and risk assessment.
The paper argues that frontier AI companies should report both pre- and post-mitigation safety evaluations to inform policy decisions, showing that relying on either alone can mislead about model safety, and identifies gaps such as lack of standardization and vague results in current disclosures.
The rapid advancement of AI systems has raised widespread concerns about potential harms of frontier AI systems and the need for responsible evaluation and oversight. In this position paper, we argue that frontier AI companies should report both pre- and post-mitigation safety evaluations to enable informed policy decisions. Evaluating models at both stages provides policymakers with essential evidence to regulate deployment, access, and safety standards. We show that relying on either in isolation can create a misleading picture of model safety. Our analysis of AI safety disclosures from leading frontier labs identifies three critical gaps: (1) companies rarely evaluate both pre- and post-mitigation versions, (2) evaluation methods lack standardization, and (3) reported results are often too vague to inform policy. To address these issues, we recommend mandatory disclosure of pre- and post-mitigation capabilities to approved government bodies, standardized evaluation methods, and minimum transparency requirements for public safety reporting. These ensure that policymakers and regulators can craft targeted safety measures, assess deployment risks, and scrutinize companies' safety claims effectively.