Large Language Models in Cybersecurity: Threats, Exposure and Mitigation

This page presents the Springer book “Large Language Models in Cybersecurity – Threats, Exposure and Mitigation” published in June, 2024. Podcasts summarizing the main findings of the book are available below the table.

Table of all the Book’s Chapter

Book Chapter	Update	Podcast
Introduction
From Deep Neural Language Models to LLMs	Overleaf	YouTube
Adapting LLMs to Downstream Applications	Overleaf	YouTube
Overview of Existing LLM Families	Overleaf	YouTube
Conversational Agents	Overleaf
Fundamental Limitations of Generative LLMs	Overleaf	YouTube
Tasks for LLMs and Their Evaluation	Overleaf	YouTube
LLMs in Cybersecurity
Private Information Leakage in LLMs	Overleaf	YouTube
Phishing and Social Engineering in the Age of LLMs	Overleaf	YouTube
Vulnerabilities Introduced by LLMs Through Code Suggestions	Overleaf	YouTube
LLM Controls Execution Flow Hijacking	Overleaf	YouTube
LLM-Aided Social Media Influence Operations	Overleaf
Deep(er) Web Indexing with LLMs	Overleaf	YouTube
Tracking and Forecasting Exposure
LLM Adoption Trends and Associated Risks	Overleaf
The Flow of Investments in the LLM Space	Overleaf	YouTube
Insurance Outlook for LLM-Induced Risk	Overleaf	YouTube
Copyright-Related Risks in the Creation and Use of ML/AI Systems	Overleaf
Monitoring Emerging Trends in LLM Research	Overleaf
Mitigation
Enhancing Security Awareness and Education for LLMs	Overleaf	YouTube
Towards Privacy Preserving LLMs Training	Overleaf	YouTube
Adversarial Evasion on LLMs	Overleaf
Robust and Private Federated Learning on LLMs	Overleaf
LLM Detectors	Overleaf	YouTube
On-Site Deployment of LLMs	Overleaf
LLMs Red Teaming	Overleaf
Standards for LLM Security	Overleaf
Conclusion
Exploring the Dual Role of LLMs in Cybersecurity: Threats and Defenses	Overleaf
Towards Safe LLMs Integration	Overleaf	YouTube