Attacks on Third-Party APIs of Large Language Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913330062426112 |
|---|---|
| author | Zhao, Wanru Khazanchi, Vidit Xing, Haodi He, Xuanli Xu, Qiongkai Lane, Nicholas Donald |
| author_facet | Zhao, Wanru Khazanchi, Vidit Xing, Haodi He, Xuanli Xu, Qiongkai Lane, Nicholas Donald |
| contents | Large language model (LLM) services have recently begun offering a plugin ecosystem to interact with third-party API services. This innovation enhances the capabilities of LLMs, but it also introduces risks, as these plugins developed by various third parties cannot be easily trusted. This paper proposes a new attacking framework to examine security and safety vulnerabilities within LLM platforms that incorporate third-party services. Applying our framework specifically to widely used LLMs, we identify real-world malicious attacks across various domains on third-party APIs that can imperceptibly modify LLM outputs. The paper discusses the unique challenges posed by third-party API integration and offers strategic possibilities to improve the security and safety of LLM ecosystems moving forward. Our code is released at https://github.com/vk0812/Third-Party-Attacks-on-LLMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_16891 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Attacks on Third-Party APIs of Large Language Models Zhao, Wanru Khazanchi, Vidit Xing, Haodi He, Xuanli Xu, Qiongkai Lane, Nicholas Donald Cryptography and Security Artificial Intelligence Computation and Language Computers and Society Large language model (LLM) services have recently begun offering a plugin ecosystem to interact with third-party API services. This innovation enhances the capabilities of LLMs, but it also introduces risks, as these plugins developed by various third parties cannot be easily trusted. This paper proposes a new attacking framework to examine security and safety vulnerabilities within LLM platforms that incorporate third-party services. Applying our framework specifically to widely used LLMs, we identify real-world malicious attacks across various domains on third-party APIs that can imperceptibly modify LLM outputs. The paper discusses the unique challenges posed by third-party API integration and offers strategic possibilities to improve the security and safety of LLM ecosystems moving forward. Our code is released at https://github.com/vk0812/Third-Party-Attacks-on-LLMs. |
| title | Attacks on Third-Party APIs of Large Language Models |
| topic | Cryptography and Security Artificial Intelligence Computation and Language Computers and Society |
| url | https://arxiv.org/abs/2404.16891 |