AI Safety Frameworks Should Include Procedures for Model Access Decisions

TL;DR

This paper advocates for AI safety frameworks to incorporate transparent, empirical procedures for making model access decisions, aiming to better manage risks associated with different user groups and access levels.

Contribution

It introduces Responsible Access Policies (RAPs), a structured approach for AI companies to evaluate and regulate model access based on empirical capability assessments and risk profiles.

Findings

Proposes RAPs as a new framework for model access decisions

Highlights the need for empirical evaluation of model capabilities

Emphasizes risk assessment for different user categories

Abstract

The downstream use cases, benefits, and risks of AI models depend significantly on what sort of access is provided to the model, and who it is provided to. Though existing safety frameworks and AI developer usage policies recognise that the risk posed by a given model depends on the level of access provided to a given audience, the procedures they use to make decisions about model access are ad hoc, opaque, and lacking in empirical substantiation. This paper consequently proposes that frontier AI companies build on existing safety frameworks by outlining transparent procedures for making decisions about model access, which we term Responsible Access Policies (RAPs). We recommend that, at a minimum, RAPs should include the following: i) processes for empirically evaluating model capabilities given different styles of access, ii) processes for assessing the risk profiles of different categories of user, and iii) clear and robust pre-commitments regarding when to grant or revoke specific types of access for particular groups under specified conditions.