Patch Tuesday, Exploit Tuesday
Benchmarking n-day exploit generation.
•
ai security benchmarks n-day offsec
writings
Filter by Tag
No Free Lunches
How do you prevent a big lab from eating your lunch?
•
ai
We Are Not Magicians
Why automating offensive operations requires engineering decision-making systems, not just giving LLMs tools
•
offsec ai ontology automation
smoltalk: RCE in Open Source Agents
A vulnerability analysis of RCE in open source AI agents through prompt injection and dangerous imports
•
security ai rce prompt-injection
Curmudgeons
Reflections on the security industry's relationship with technology and AI
•
opinion ai security
Why LLM Security Is Important
Exploring the real-world security implications of Large Language Models and AI systems
•
security ai llm