Abstract
We surface a new threat to closed-weight Large Language Models (LLMs) that enables an attacker to compute optimization-based prompt injections. Specifically, we characterize how an attacker can leverage the loss-like information returned from the remote fine-tuning interface to guide the search for adversarial prompts. The fine-tuning interface is hosted by an LLM vendor and allows developers to fine-tune LLMs for their tasks, thus providing utility, but also exposes enough information for an attacker to compute adversarial prompts. Through an experimental analysis, we characterize the loss-like values returned by the Gemini fine-tuning API and demonstrate that they provide a useful signal for discrete optimization of adversarial prompts using a greedy search algorithm. Using the PurpleLlama prompt injection benchmark, we demonstrate attack success rates between 65% and 82% on Google's Gemini family of LLMs. These attacks exploit the classic utility-security tradeoff - the fine-tuning interface provides a useful feature for developers but also exposes the LLMs to powerful attacks.
Impact
Google Gemini's Patch:
We constrained the API parameters that they were relying on. In particular, capping the learning rate to a value that would rule out small perturbations and limiting the batch size to a minimum of 4, such that they can no longer correlate the reported loss values to the individual inputs.
Media coverage: Arstechnica and Andriod Authority
Find this paper at https://arxiv.org/abs/2501.09798 and https://huggingface.co/papers/2501.09798!