Assessing Security and Privacy Concerns in LLM Applications: Legal and Social Concerns

Title

Assessing Security and Privacy Concerns in LLM Applications: Legal and Social Concerns

Creator

Singh, Bhupinder; Shastri, Arunima; Chandra, Saurabh

Description

Large language models (LLMs) like Open AI's GPT- 4 have transformed many industries, including natural language processing. But, still the applications of these need a properly secure mechanism as it is getting attached to everyday technology at on an extensive rate and large security & privacy risk are encapsulated in its use which must be controlled for safe and ethical use. LLM models are trained with an avalanche of data, often from web- scraped sources that could be handling personal and sensitive content. This may inadvertently have the LAZ model hold on to, and possibly leak, some private data of users who use the API. Having measures to address them through comprehensive strategies and conforming with regulations are important in the responsible usage of LLM applications. 2026 by IGI Global Scientific Publishing. All rights reserved.

Source

Theory, Practice, and Future Direction of Large Language Models;pp.227-249

Date

01-01-2025

Publisher

IGI Global

Coverage

Singh B., Sharda University, India; Shastri A., Christ University, Bangalore, India; Chandra S., Bennett University, Greater Noida, India

Rights

Restricted Access; Hardcopy may be available in the library

Relation

ISBN: 979-836938389-6; 979-836938387-2;

Format

online

Language

English

Type

Book chapter

Identifier

https://doi.org/10.4018/979-8-3693-8387-2.ch008

https://www.scopus.com/pages/publications/105021208470?origin=resultslist