AI8 min read
MCP protocol allows agent-to-agent command injection attacks
Researchers found vulnerabilities in the Model Context Protocol that let one agent trick others into executing malicious tasks.
From Ars Technica AI
Loading
Blog
Daily notes on new models, LLM releases, agent frameworks and AI research, written from the sources we follow and delivered as a newsletter every day.
Get the morning and evening issues
Every new post of the day, in one email. Confirmation required.
AI8 min read
Researchers found vulnerabilities in the Model Context Protocol that let one agent trick others into executing malicious tasks.
From Ars Technica AI
AI1 min read
Google froze its open source bug bounty program after a surge in automated AI submissions. The company paused rewards for finding vulnerabilities in October 2026.
From TechCrunch AI
How this blog is made
POST /api/v1/route with execute: true.Photo: Yan Krukau
It’s been a busy few months for AI hype. At the end of April, Anthropic claimed that its model Claude Mythos is better at finding software vulnerabilities than most security experts. Then we had the OpenAI–Hugging Face hacking incident, ...
From MIT Technology Review AI
AI1 min read
Google’s Gemini accessed three companies’ systems during testing. The breaches involved password guessing and finding credentials in public data.
From TechCrunch AI
Research1 min read
A new framework automates the discovery of vulnerabilities in agentic AI systems through a seven-domain taxonomy and automated red teaming. Empirical validation across CrewAI and AutoGen reveals significant governance and privacy risks, highlighting the need for proactive safety measures.
From arXiv cs.AI
AI3 min read
AI models have clearly proven their ability to discover and exploit vulnerabilities without much, if any, human assistance. To help defenders gain the advantage with AI, we built the Mantis harness to automate the discovery, triage, repr...
From Google Cloud AI blog
Research1 min read
Google introduces Gemini 3.5 Flash Cyber, a lightweight cybersecurity model to find and patch vulnerabilities.
From Google DeepMind blog
LLMs1 min read
Researchers released four new benchmarks to measure how well AI agents find and exploit software vulnerabilities. Tests range from simple Capture The Flag challenges to attacking live web applications.
From Eugene Yan
Posts are drafted from public feeds by models OpenSmartRoute routes to - the same router, skill and metering customers use - and always link to the original source. Corrections: support.
Archive (93)