Weaponizing LLMs to Hack JavaScript AI Applications
watch on portal.gitnation.org ↗https://portal.gitnation.org/contents/weaponizing-llms-to-hack-javascript-ai-applicationssummary
Covers risks from using LLM-based code assistants and integrating LLMs into JavaScript/AI app workflows, showing how generative AI can introduce insecure code (path traversal, prototype pollution, SQL injection) and how attackers can use prompt injection to bypass controls and weaponize LLMs for SQL injection and business-logic attacks. Demonstrations use real-world application hacking to show insecure patterns and then discusses using AI security guardrails to secure generative-AI code.
AI-written summary · about
related talks
26mWhen AI Gets Manipulated: Security Risks in LLM Applications
26mHow Bad Actors AbUse LLMs The Mechanisms That Enable Misuse and What We Can Do About
39mLLMs everywhere: The future is now and the past keeps repeating itself
36mCompromising LLMs: The Advent of AI Malware
31mHacking AI Applications
40mFrom Text to Flaws: Application Vulnerabilities with Generative AI and LLMs
43mSecurity of GenAI & LLMs: A New Era of Ethical Hacking
43mLLMs in a Hacker's Hands: From Prompt Injection to Data Leakage