Tag:
vulnerabilities
AI
OpenAI Aardvark: Automated Vulnerability Detection Made Easy
OpenAI has launched Aardvark, a groundbreaking autonomous security agent designed to detect and resolve code vulnerabilities, now available in a private beta. Aardvark boasts...
AI Hacker News
Exploiting Prompt Injection for Remote Code Execution in AI Agents – Insights from the Trail of Bits Blog
Uncovering Argument Injection in AI: A Deep Dive
In the evolving landscape of artificial intelligence, security vulnerabilities in AI agents reveal crucial design antipatterns that...
AI Hacker News
Exploiting Image Scaling Vulnerabilities in Production AI Systems – The Trail of Bits Blog
🚨 Unlocking AI Vulnerabilities: The Dark Side of Image Scaling 🚨
In our latest blog post, we reveal a groundbreaking method that exploits image scaling...
AI
How One Malicious Document Could Expose Confidential Data Through ChatGPT
The latest generative AI models, like OpenAI’s ChatGPT, can connect to personal data systems, such as Gmail, GitHub, and Microsoft Calendar, to deliver tailored...
AI Hacker News
Cursor’s AI Coding Agent Transformed into Local Shell with a Simple One-Line Prompt Exploit
🚨 AI Security Alert: Vulnerability Exposed in Cursor Code Editing Software 🚨
Recently, AimLabs uncovered a significant data-poisoning attack that impacts Cursor, an AI-driven code...
AI
The Emergence of ‘Vibe Hacking’: A New AI Challenge on the Horizon
In 2023, Trend Micro researchers demonstrated that ChatGPT could generate malicious code by simulating a security researcher role. This capability poses a threat, particularly...