Using Large Language Models with Sensitive Company Data

Recent developments in the area of large language models (LLMs), especially domain adaption and the integration with different data sources, have significantly expanded their capabilities and applications.

Our research aims to explore various strategies for companies to utilize LLMs with company data, with a special focus on ensuring data protection, adherence to authorization policies, and compliance with governmental regulations. This exploration will primarily be conducted through interviews with industry experts to gather insights on different architectures and technologies that safeguard data privacy and security.

The goal is to provide actionable insights by summarizing the findings from industry interviews, highlighting successful case studies, challenges, and future trends in the use of LLMs with proprietary company data. The research team includes Burkhardt Funk, Paul Drews, and Jann Pfeifer (Leuphana).

We would be honored if you could participate as an interviewee. If you are interested, please contact us at jann.pfeifer@stud.leuphana.de. We will ensure that the findings and insights from the interview are shared in an anonymized format.

Further reading:

• Minaee, S., Mikolov, T., Nikzad, N., Chenaghlu, M., Socher, R., Amatriain, X., & Gao, J. (2024). Large Language Models: A Survey. arXiv [Cs.CL]. https://arxiv.org/abs/2402.06196

• Sun, L., Huang, Y., Wang, H., Wu, S., Zhang, Q., Gao, C., … Zhao, Y. (2024). TrustLLM: Trustworthiness in Large Language Models. arXiv [Cs.CL]. Retrieved from https://arxiv.org/abs/2401.05561

• Yao, Y., Duan, J., Xu, K., Cai, Y., Sun, Z., & Zhang, Y. (2024). A Survey on Large Language Model (LLM) Security and Privacy: The Good, the Bad, and the Ugly. arXiv [Cs.CR]. https://arxiv.org/abs/2312.02003

• Ling, C., Zhao, X., Lu, J., Deng, C., Zheng, C., Wang, J., … Zhao, L. (2023). Domain Specialization as the Key to Make Large Language Models Disruptive: A Comprehensive Survey. arXiv [ Cs.CL]. https://arxiv.org/abs/2305.18703

• Guo, X., & Yu, H. (2022). On the Domain Adaptation and Generalization of Pretrained Language Models: A Survey. arXiv [Cs.CL]. https://arxiv.org/abs/2211.03154

• Namer, Assaf; Miller, Jim; Kulkarni, Prashant; Vagts, Hauke; Bisson, Jason; and Maltzman, Brandon, "Tenant Data Security for LLM Applications in a Multi-Tenancy Environment", Technical Disclosure Commons, (January 12, 2024). https://www.tdcommons.org/dpubs_series/6596