Hazards from Increasingly Accessible Fine-Tuning of Downloadable Foundation Models

TL;DR

The paper discusses how the increased accessibility of fine-tuning foundation models can lead to greater hazards, emphasizing the need for mitigation strategies amidst uncertain risks.

Contribution

It analyzes the implications of accessible fine-tuning, highlighting hazards, potential mitigations, and the balance of benefits and risks.

Findings

Accessible fine-tuning may facilitate malicious use

It complicates oversight of dangerous capabilities

Mitigation measures are urgently needed

Abstract

Public release of the weights of pretrained foundation models, otherwise known as downloadable access \citep{solaiman_gradient_2023}, enables fine-tuning without the prohibitive expense of pretraining. Our work argues that increasingly accessible fine-tuning of downloadable models may increase hazards. First, we highlight research to improve the accessibility of fine-tuning. We split our discussion into research that A) reduces the computational cost of fine-tuning and B) improves the ability to share that cost across more actors. Second, we argue that increasingly accessible fine-tuning methods may increase hazard through facilitating malicious use and making oversight of models with potentially dangerous capabilities more difficult. Third, we discuss potential mitigatory measures, as well as benefits of more accessible fine-tuning. Given substantial remaining uncertainty about hazards, we conclude by emphasizing the urgent need for the development of mitigations.