Assessing Language Model Deployment with Risk Cards

TL;DR

<3-5 sentence high-level summary> The paper addresses the lack of a risk-centric, deployment-context framework for language-model harms by introducing RiskCards. RiskCards provide a structured card-based format to document risk pathways, affected actors, and concrete prompt-output examples, decoupled from any single model or dataset. The authors outline an end-to-end workflow for defining assessments, selecting risk cards, provisioning assessors, and probing models, emphasizing qualitative human evaluation over automated benchmarks. A starter set of RiskCards is presented, drawn from a literature survey, along with considerations on sustainability, governance, and potential dual-use concerns.

Abstract

This paper introduces RiskCards, a framework for structured assessment and documentation of risks associated with an application of language models. As with all language, text generated by language models can be harmful, or used to bring about harm. Automating language generation adds both an element of scale and also more subtle or emergent undesirable tendencies to the generated text. Prior work establishes a wide variety of language model harms to many different actors: existing taxonomies identify categories of harms posed by language models; benchmarks establish automated tests of these harms; and documentation standards for models, tasks and datasets encourage transparent reporting. However, there is no risk-centric framework for documenting the complexity of a landscape in which some risks are shared across models and contexts, while others are specific, and where certain conditions may be required for risks to manifest as harms. RiskCards address this methodological gap by providing a generic framework for assessing the use of a given language model in a given scenario. Each RiskCard makes clear the routes for the risk to manifest harm, their placement in harm taxonomies, and example prompt-output pairs. While RiskCards are designed to be open-source, dynamic and participatory, we present a "starter set" of RiskCards taken from a broad literature survey, each of which details a concrete risk presentation. Language model RiskCards initiate a community knowledge base which permits the mapping of risks and harms to a specific model or its application scenario, ultimately contributing to a better, safer and shared understanding of the risk landscape.

Figures (1)

• Figure 1: Overview of proposed risk cards.