Enhancing user prompt confidentiality in Large Language Models through advanced differential encryption

被引:3
|
作者
Gupta, Brij B. [1 ,2 ,3 ,4 ,5 ]
Gaurav, Akshat [6 ]
Arya, Varsha [7 ,8 ]
Alhalabi, Wadee [9 ]
Alsalman, Dheyaaldin [10 ]
Vijayakumar, P. [11 ]
机构
[1] Asia Univ, Int Ctr AI & Cyber Secur Res & Innovat CCRI, Taichung, Taiwan
[2] Asia Univ, Dept Comp Sci & Informat Engn, Taichung, Taiwan
[3] Kyung Hee Univ, 26 Kyungheedae Ro, Seoul, South Korea
[4] Symbiosis Int Univ, Symbiosis Ctr Informat Technol SCIT, Pune, India
[5] Univ Petr & Energy Studies UPES, Ctr Interdisciplinary Res, Dehra Dun, India
[6] Ronin Inst, Montclair, NJ USA
[7] Asia Univ, Dept Business Adm, Taichung, Taiwan
[8] Lebanese Amer Univ, Dept Elect & Comp Engn, Beirut 1102, Lebanon
[9] King Abdulaziz Univ, Dept Comp Sci, Immers Virtual Real Res Grp, Jeddah, Saudi Arabia
[10] Dar Al Hekma Univ, Sch Engn Comp & Informat, Jeddah, Saudi Arabia
[11] Univ Coll Engn Tindivanam, Dept Comp Sci & Engn, Tindivanam 604001, Tamil Nadu, India
关键词
Cryptographic privacy; Large Language Models; Data anonymization; Secure AI framework; Personal data protection; AUTHENTICATION PROTOCOL; DESIGN;
D O I
10.1016/j.compeleceng.2024.109215
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
In the era of artificial intelligence (AI) advancements heralded by Large Language Models (LLMs) like GPT-3, the capacity to parse and generate human -like text brings to light substantial privacy concerns. These arise notably from LLMs' reliance on vast datasets often laden with personal information, underscoring the potential for inadvertent memorization and disclosure of sensitive data. Addressing these pivotal privacy concerns, our research introduces a novel two -fold approach aimed at bolstering the confidentiality and security of user data in LLM applications. Firstly, we deploy advanced cryptographic techniques, incorporating bespoke encryption and hashing protocols, to preprocess user data. This strategy effectively anonymizes personal identifiers prior to their processing by LLMs, directly tackling the challenges of sensitive information exposure. Concurrently, our methodology encompasses a secure mutual authentication protocol utilizing lightweight cryptographic measures. This ensures that system interactions are strictly reserved for authenticated users, thereby enhancing overall data security. Collectively, our approach not only preserves the utility of data for AI tasks but also fortifies the privacy framework surrounding LLMs, significantly reducing the likelihood of privacy breaches and steering AI development towards a more secure and ethically grounded future.
引用
收藏
页数:13
相关论文
共 50 条
  • [21] Towards Taming Large Language Models with Prompt Templates for Legal GRL Modeling
    de Kinderen, Sybren
    Winter, Karolin
    ENTERPRISE, BUSINESS-PROCESS AND INFORMATION SYSTEMS MODELING, BPMDS 2024, EMMSAD 2024, 2024, 511 : 213 - 228
  • [22] Turning Large Language Models into AI Assistants for Startups Using Prompt Patterns
    Wang, Xiaofeng
    Attal, Mohammad Idris
    Rafiq, Usman
    Hubner-Benz, Sylvia
    AGILE PROCESSES IN SOFTWARE ENGINEERING AND EXTREME PROGRAMMING - WORKSHOPS, XP 2022 WORKSHOPS, XP 2023 WORKSHOPS, 2024, 489 : 192 - 200
  • [23] Trend Analysis Through Large Language Models
    Alzapiedi, Lucas
    Bihl, Trevor
    IEEE NATIONAL AEROSPACE AND ELECTRONICS CONFERENCE, NAECON 2024, 2024, : 370 - 374
  • [24] Leveraging Large Language Models for Enhancing Safety in Maritime Operations
    Miller, Tymoteusz
    Durlik, Irmina
    Kostecka, Ewelina
    Lobodzinska, Adrianna
    Lazuga, Kinga
    Kozlovska, Polina
    APPLIED SCIENCES-BASEL, 2025, 15 (03):
  • [25] GenKP: generative knowledge prompts for enhancing large language models
    Li, Xinbai
    Peng, Shaowen
    Yada, Shuntaro
    Wakamiya, Shoko
    Aramaki, Eiji
    APPLIED INTELLIGENCE, 2025, 55 (06)
  • [26] Enhancing Legal Compliance and Regulation Analysis with Large Language Models
    Hassani, Shabnam
    32ND IEEE INTERNATIONAL REQUIREMENTS ENGINEERING CONFERENCE, RE 2024, 2024, : 507 - 511
  • [27] Enhancing the accessibility of regionalization techniques through large language models: a case study in conversational agent guidance
    Feng, Xin
    Cao, Yuanpei
    INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE, 2024,
  • [28] Enhancing Genetic Improvement Mutations Using Large Language Models
    Brownlee, Alexander E. I.
    Callan, James
    Even-Mendoza, Karine
    Geiger, Alina
    Hanna, Carol
    Petke, Justyna
    Sarro, Federica
    Sobania, Dominik
    SEARCH-BASED SOFTWARE ENGINEERING, SSBSE 2023, 2024, 14415 : 153 - 159
  • [29] Evaluating and enhancing spatial cognition abilities of large language models
    Yang, Anran
    Fu, Cheng
    Jia, Qingren
    Dong, Weihua
    Ma, Mengyu
    Chen, Hao
    Yang, Fei
    Wu, Hui
    INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE, 2025,
  • [30] Enhancing Translation Validation of Compiler Transformations with Large Language Models
    Wang, Yanzhao
    Xie, Fei
    INTERNATIONAL JOURNAL OF SOFTWARE ENGINEERING AND KNOWLEDGE ENGINEERING, 2025, 35 (01) : 45 - 57