From the course: A Bug Bounty Toolkit for Security Researchers
Join today to access over 24,800 courses taught by industry experts.
Leveraging LLM for security testing
From the course: A Bug Bounty Toolkit for Security Researchers
Leveraging LLM for security testing
- [Presenter] The rise of AI and machine learning has transformed many industries, including cybersecurity. Staying ahead of evolving AI threats require continuous learning and adaptation. Keep updated with latest AI trends, research, and threat intelligence. Understanding the capabilities and limitation of AI can help you anticipate and respond to emerging threats. Researchers have been studying large language models like ChatGPT-4 for the capabilities and limitations, but have also identified vulnerabilities and security challenges associated with these advanced models. Let's look at some recent vulnerabilities found in LLMs. First is prompt injection. Prompt injection occurs when attackers craft input that manipulate the model's behavior or output by injecting malicious prompts, leading the model to execute unintended task, bypass restriction, or generate harmful content. For example, if your model is instructed, do not respond to any input with sensitive information An attacker…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.