LinkedInGuard: A Personal Web-Based AI Security Assistant for Detecting Threats

Document Type

Conference Proceeding

Source of Publication

2026 17th Student Research Conference on Applied Computing Src 2026

Publication Date

4-22-2026

Abstract

The rapid growth of professional networking platforms such as LinkedIn has increased users' exposure to cybersecurity threats, including phishing, fake job offers, impersonation, and malicious content. While existing platform defenses are often reactive or designed for organizational use, there is a lack of accessible tools that help individual users evaluate suspicious content before becoming victims. This paper presents LinkedInGuard, an artificial intelligence-driven web application that leverages open-source large language models to analyze user-submitted LinkedIn text, posts, or files and classify potential risks such as phishing, fake job offers, and other suspicious activity. The system presents results through a user-friendly dashboard that explains risks and suggests safer actions. Full background automation was not implemented due to platform access restrictions and associated costs, which were beyond the scope of this student project. Experimental testing shows that LinkedInGuard can effectively assist users in identifying harmful content and improving their online decision-making, contributing to enhanced trust, awareness, and digital safety on professional networking platforms.

ISBN

[9798319510167]

Publisher

IEEE

Disciplines

Computer Sciences

Keywords

cybersecurity, fake profiles, LinkedIn security, machine learning, social engineering

Scopus ID

105043183941

Indexed in Scopus

yes

Open Access

no

Share

COinS