Edoardo Debenedetti

Edoardo Debenedetti, Giorgio Severi, Nicholas Carlini et al. · berkeley, deepmind

Most current approaches for protecting privacy in machine learning (ML) assume that models exist in a vacuum. Yet, in reality, these models are part of larger systems that include components for training data filtering, output monitoring, and more. In this work, we introduce privacy side channels: attacks that exploit these system-level components to extract private information at far higher rates than is otherwise possible for standalone models. We propose four categories of side channels that span the entire ML lifecycle (training data filtering, input preprocessing, output post-processing, and query filtering) and allow for enhanced membership inference, data extraction, and even novel threats such as extraction of users' test queries. For example, we show that deduplicating training data before applying differentially-private training creates a side-channel that completely invalidates any provable privacy guarantees. We further show that systems which block language models from regenerating training data can be exploited to exfiltrate private keys contained in the training set--even if the model did not memorize these keys. Taken together, our results demonstrate the need for a holistic, end-to-end privacy analysis of machine learning systems.

Edoardo Debenedetti, Vikash Sehwag, Prateek Mittal · eth-zurich, princeton

In this paper, we ask whether Vision Transformers (ViTs) can serve as an underlying architecture for improving the adversarial robustness of machine learning models against evasion attacks. While earlier works have focused on improving Convolutional Neural Networks, we show that also ViTs are highly suitable for adversarial training to achieve competitive performance. We achieve this objective using a custom adversarial training recipe, discovered using rigorous ablation studies on a subset of the ImageNet dataset. The canonical training recipe for ViTs recommends strong data augmentation, in part to compensate for the lack of vision inductive bias of attention modules, when compared to convolutions. We show that this recipe achieves suboptimal performance when used for adversarial training. In contrast, we find that omitting all heavy data augmentation, and adding some additional bag-of-tricks ($\varepsilon$-warmup and larger weight decay), significantly boosts the performance of robust ViTs. We show that our recipe generalizes to different classes of ViT architectures and large-scale models on full ImageNet-1k. Additionally, investigating the reasons for the robustness of our models, we show that it is easier to generate strong attacks during training when using our recipe and that this leads to better robustness at test time. Finally, we further study one consequence of adversarial training by proposing a way to quantify the semantic nature of adversarial perturbations and highlight its correlation with the robustness of the model. Overall, we recommend that the community should avoid translating the canonical training recipes in ViTs to robust training and rethink common training choices in the context of adversarial training.

Edoardo Debenedetti, Nicholas Carlini, Florian Tramèr · eth-zurich

Decision-based evasion attacks repeatedly query a black-box classifier to generate adversarial examples. Prior work measures the cost of such attacks by the total number of queries made to the classifier. We argue this metric is flawed. Most security-critical machine learning systems aim to weed out "bad" data (e.g., malware, harmful content, etc). Queries to such systems carry a fundamentally asymmetric cost: queries detected as "bad" come at a higher cost because they trigger additional security filters, e.g., usage throttling or account suspension. Yet, we find that existing decision-based attacks issue a large number of "bad" queries, which likely renders them ineffective against security-critical systems. We then design new attacks that reduce the number of bad queries by $1.5$-$7.3\times$, but often at a significant increase in total (non-bad) queries. We thus pose it as an open problem to build black-box attacks that are more effective under realistic cost metrics.

Patrick Chao, Edoardo Debenedetti, Alexander Robey et al. · eth-zurich, princeton

Jailbreak attacks cause large language models (LLMs) to generate harmful, unethical, or otherwise objectionable content. Evaluating these attacks presents a number of challenges, which the current collection of benchmarks and evaluation techniques do not adequately address. First, there is no clear standard of practice regarding jailbreaking evaluation. Second, existing works compute costs and success rates in incomparable ways. And third, numerous works are not reproducible, as they withhold adversarial prompts, involve closed-source code, or rely on evolving proprietary APIs. To address these challenges, we introduce JailbreakBench, an open-sourced benchmark with the following components: (1) an evolving repository of state-of-the-art adversarial prompts, which we refer to as jailbreak artifacts; (2) a jailbreaking dataset comprising 100 behaviors -- both original and sourced from prior work (Zou et al., 2023; Mazeika et al., 2023, 2024) -- which align with OpenAI's usage policies; (3) a standardized evaluation framework at https://github.com/JailbreakBench/jailbreakbench that includes a clearly defined threat model, system prompts, chat templates, and scoring functions; and (4) a leaderboard at https://jailbreakbench.github.io/ that tracks the performance of attacks and defenses for various LLMs. We have carefully considered the potential ethical implications of releasing this benchmark, and believe that it will be a net positive for the community.

Edoardo Debenedetti, Ilia Shumailov, Tianqi Fan et al. · deepmind, eth-zurich

Large Language Models (LLMs) are increasingly deployed in agentic systems that interact with an untrusted environment. However, LLM agents are vulnerable to prompt injection attacks when handling untrusted data. In this paper we propose CaMeL, a robust defense that creates a protective system layer around the LLM, securing it even when underlying models are susceptible to attacks. To operate, CaMeL explicitly extracts the control and data flows from the (trusted) query; therefore, the untrusted data retrieved by the LLM can never impact the program flow. To further improve security, CaMeL uses a notion of a capability to prevent the exfiltration of private data over unauthorized data flows by enforcing security policies when tools are called. We demonstrate effectiveness of CaMeL by solving $77\%$ of tasks with provable security (compared to $84\%$ with an undefended system) in AgentDojo. We release CaMeL at https://github.com/google-research/camel-prompt-injection.

Nicholas Carlini, Javier Rando, Edoardo Debenedetti et al. · eth-zurich

We introduce AutoAdvExBench, a benchmark to evaluate if large language models (LLMs) can autonomously exploit defenses to adversarial examples. Unlike existing security benchmarks that often serve as proxies for real-world tasks, bench directly measures LLMs' success on tasks regularly performed by machine learning security experts. This approach offers a significant advantage: if a LLM could solve the challenges presented in bench, it would immediately present practical utility for adversarial machine learning researchers. We then design a strong agent that is capable of breaking 75% of CTF-like ("homework exercise") adversarial example defenses. However, we show that this agent is only able to succeed on 13% of the real-world defenses in our benchmark, indicating the large gap between difficulty in attacking "real" code, and CTF-like code. In contrast, a stronger LLM that can attack 21% of real defenses only succeeds on 54% of CTF-like defenses. We make this benchmark available at https://github.com/ethz-spylab/AutoAdvExBench.

Edoardo Debenedetti, Jie Zhang, Mislav Balunović et al.

AI agents aim to solve complex tasks by combining text-based reasoning with external tool calls. Unfortunately, AI agents are vulnerable to prompt injection attacks where data returned by external tools hijacks the agent to execute malicious tasks. To measure the adversarial robustness of AI agents, we introduce AgentDojo, an evaluation framework for agents that execute tools over untrusted data. To capture the evolving nature of attacks and defenses, AgentDojo is not a static test suite, but rather an extensible environment for designing and evaluating new agent tasks, defenses, and adaptive attacks. We populate the environment with 97 realistic tasks (e.g., managing an email client, navigating an e-banking website, or making travel bookings), 629 security test cases, and various attack and defense paradigms from the literature. We find that AgentDojo poses a challenge for both attacks and defenses: state-of-the-art LLMs fail at many tasks (even in the absence of attacks), and existing prompt injection attacks break some security properties but not all. We hope that AgentDojo can foster research on new design principles for AI agents that solve common tasks in a reliable and robust manner.. We release the code for AgentDojo at https://github.com/ethz-spylab/agentdojo.

Edoardo Debenedetti, Javier Rando, Daniel Paleka et al.

Large language model systems face important security risks from maliciously crafted messages that aim to overwrite the system's original instructions or leak private data. To study this problem, we organized a capture-the-flag competition at IEEE SaTML 2024, where the flag is a secret string in the LLM system prompt. The competition was organized in two phases. In the first phase, teams developed defenses to prevent the model from leaking the secret. During the second phase, teams were challenged to extract the secrets hidden for defenses proposed by the other teams. This report summarizes the main insights from the competition. Notably, we found that all defenses were bypassed at least once, highlighting the difficulty of designing a successful defense and the necessity for additional research to protect LLM systems. To foster future research in this direction, we compiled a dataset with over 137k multi-turn attack chats and open-sourced the platform.

Francesco Croce, Maksym Andriushchenko, Vikash Sehwag et al.

As a research community, we are still lacking a systematic understanding of the progress on adversarial robustness which often makes it hard to identify the most promising ideas in training robust models. A key challenge in benchmarking robustness is that its evaluation is often error-prone leading to robustness overestimation. Our goal is to establish a standardized benchmark of adversarial robustness, which as accurately as possible reflects the robustness of the considered models within a reasonable computational budget. To this end, we start by considering the image classification task and introduce restrictions (possibly loosened in the future) on the allowed models. We evaluate adversarial robustness with AutoAttack, an ensemble of white- and black-box attacks, which was recently shown in a large-scale study to improve almost all robustness evaluations compared to the original publications. To prevent overadaptation of new defenses to AutoAttack, we welcome external evaluations based on adaptive attacks, especially where AutoAttack flags a potential overestimation of robustness. Our leaderboard, hosted at https://robustbench.github.io/, contains evaluations of 120+ models and aims at reflecting the current state of the art in image classification on a set of well-defined tasks in $\ell_\infty$- and $\ell_2$-threat models and on common corruptions, with possible extensions in the future. Additionally, we open-source the library https://github.com/RobustBench/robustbench that provides unified access to 80+ robust models to facilitate their downstream applications. Finally, based on the collected models, we analyze the impact of robustness on the performance on distribution shifts, calibration, out-of-distribution detection, fairness, privacy leakage, smoothness, and transferability.

Joshua Freeman, Chloe Rippe, Edoardo Debenedetti et al. · eth-zurich

Copyright infringement in frontier LLMs has received much attention recently due to the New York Times v. OpenAI lawsuit, filed in December 2023. The New York Times claims that GPT-4 has infringed its copyrights by reproducing articles for use in LLM training and by memorizing the inputs, thereby publicly displaying them in LLM outputs. Our work aims to measure the propensity of OpenAI's LLMs to exhibit verbatim memorization in its outputs relative to other LLMs, specifically focusing on news articles. We discover that both GPT and Claude models use refusal training and output filters to prevent verbatim output of the memorized articles. We apply a basic prompt template to bypass the refusal training and show that OpenAI models are currently less prone to memorization elicitation than models from Meta, Mistral, and Anthropic. We find that as models increase in size, especially beyond 100 billion parameters, they demonstrate significantly greater capacity for memorization. Our findings have practical implications for training: more attention must be placed on preventing verbatim memorization in very large models. Our findings also have legal significance: in assessing the relative memorization capacity of OpenAI's LLMs, we probe the strength of The New York Times's copyright infringement claims and OpenAI's legal defenses, while underscoring issues at the intersection of generative AI, law, and policy.

Luca Beurer-Kellner, Beat Buesser, Ana-Maria Creţu et al. · eth-zurich

As AI agents powered by Large Language Models (LLMs) become increasingly versatile and capable of addressing a broad spectrum of tasks, ensuring their security has become a critical challenge. Among the most pressing threats are prompt injection attacks, which exploit the agent's resilience on natural language inputs -- an especially dangerous threat when agents are granted tool access or handle sensitive information. In this work, we propose a set of principled design patterns for building AI agents with provable resistance to prompt injection. We systematically analyze these patterns, discuss their trade-offs in terms of utility and security, and illustrate their real-world applicability through a series of case studies.

Michael Aerni, Javier Rando, Edoardo Debenedetti et al. · eth-zurich

Large language models memorize parts of their training data. Memorizing short snippets and facts is required to answer questions about the world and to be fluent in any language. But models have also been shown to reproduce long verbatim sequences of memorized text when prompted by a motivated adversary. In this work, we investigate an intermediate regime of memorization that we call non-adversarial reproduction, where we quantify the overlap between model responses and pretraining data when responding to natural and benign prompts. For a variety of innocuous prompt categories (e.g., writing a letter or a tutorial), we show that up to 15% of the text output by popular conversational language models overlaps with snippets from the Internet. In worst cases, we find generations where 100% of the content can be found exactly online. For the same tasks, we find that human-written text has far less overlap with Internet data. We further study whether prompting strategies can close this reproduction gap between models and humans. While appropriate prompting can reduce non-adversarial reproduction on average, we find that mitigating worst-case reproduction of training data requires stronger defenses -- even for benign interactions.

Edoardo Debenedetti, Zishen Wan, Maksym Andriushchenko et al. · eth-zurich, princeton

The last six years have witnessed significant progress in adversarially robust deep learning. As evidenced by the CIFAR-10 dataset category in RobustBench benchmark, the accuracy under $\ell_\infty$ adversarial perturbations improved from 44\% in \citet{Madry2018Towards} to 71\% in \citet{peng2023robust}. Although impressive, existing state-of-the-art is still far from satisfactory. It is further observed that best-performing models are often very large models adversarially trained by industrial labs with significant computational budgets. In this paper, we aim to understand: ``how much longer can computing power drive adversarial robustness advances?" To answer this question, we derive \emph{scaling laws for adversarial robustness} which can be extrapolated in the future to provide an estimate of how much cost we would need to pay to reach a desired level of robustness. We show that increasing the FLOPs needed for adversarial training does not bring as much advantage as it does for standard training in terms of performance improvements. Moreover, we find that some of the top-performing techniques are difficult to exactly reproduce, suggesting that they are not robust enough for minor changes in the training setup. Our analysis also uncovers potentially worthwhile directions to pursue in future research. Finally, we make our benchmarking framework (built on top of \texttt{timm}~\citep{rw2019timm}) publicly available to facilitate future analysis in efficient robust deep learning.

Fredrik Nestaas, Edoardo Debenedetti, Florian Tramèr

Large Language Models (LLMs) are increasingly used in applications where the model selects from competing third-party content, such as in LLM-powered search engines or chatbot plugins. In this paper, we introduce Preference Manipulation Attacks, a new class of attacks that manipulate an LLM's selections to favor the attacker. We demonstrate that carefully crafted website content or plugin documentations can trick an LLM to promote the attacker products and discredit competitors, thereby increasing user traffic and monetization. We show this leads to a prisoner's dilemma, where all parties are incentivized to launch attacks, but the collective effect degrades the LLM's outputs for everyone. We demonstrate our attacks on production LLM search engines (Bing and Perplexity) and plugin APIs (for GPT-4 and Claude). As LLMs are increasingly used to rank third-party content, we expect Preference Manipulation Attacks to emerge as a significant threat.