Removing RLHF Protections in GPT-4 via Fine-Tuning

Qiusi Zhan, Richard Fang, Rohan Bindu, Akul Gupta, Tatsunori Hashimoto, Daniel Kang

Abstract
As large language models (LLMs) have increased in their capabilities, so doestheir potential for dual use. To reduce harmful outputs, produces and vendors ofLLMs have used reinforcement learning with human feedback (RLHF). In tandem,LLM vendors have been increasingly enabling fine-tuning of their most powerfulmodels. However, concurrent work has shown that fine-tuning can remove RLHFprotections. We may expect that the most powerful models currently available(GPT-4) are less susceptible to fine-tuning attacks. In this work, we show the contrary: fine-tuning allows attackers to remove RLHFprotections with as few as 340 examples and a 95% success rate. These trainingexamples can be automatically generated with weaker models. We further show thatremoving RLHF protections does not decrease usefulness on non-censored outputs,providing evidence that our fine-tuning strategy does not decrease usefulnessdespite using weaker models to generate training data. Our results show the needfor further research on protections on LLMs.
Anthology ID:
2024.naacl-short.59
Volume:
Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers)
Month:
June
Year:
2024
Address:
Mexico City, Mexico
Editors:
Kevin Duh, Helena Gomez, Steven Bethard
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
681–687
Language:
URL:
https://aclanthology.org/2024.naacl-short.59
DOI:
Bibkey:
Cite (ACL):
Qiusi Zhan, Richard Fang, Rohan Bindu, Akul Gupta, Tatsunori Hashimoto, and Daniel Kang. 2024. Removing RLHF Protections in GPT-4 via Fine-Tuning. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers), pages 681–687, Mexico City, Mexico. Association for Computational Linguistics.
Cite (Informal):
Removing RLHF Protections in GPT-4 via Fine-Tuning (Zhan et al., NAACL 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-checklist/2024.naacl-short.59.pdf
PDF Search