AI Tools

OWASP Agent Memory Guard is a useful signal: the dangerous part of agent memory is not only what the model remembers. It is who gets to write into that memory, when, and how long the poison survives.

Google's Threat Intelligence Group confirmed the first known AI-generated zero-day exploit — a 2FA bypass built for mass exploitation. Meanwhile, state-sponsored groups are industrializing AI-powered vulnerability research.

Microsoft disclosed two critical vulnerabilities in Semantic Kernel that turn prompt injection into full remote code execution. The AI agent framework your tools are built on just became the attack surface.

Mandiant's M-Trends 2026 data shows AI in the attack chain at every stage — but the breaches still start with the same old failures. The uncomfortable truth is both things are true at once.

OpenAI opened GPT-5.4-Cyber to thousands of vetted defenders this week. Anthropic won't release Mythos publicly at all. Both models can find zero-days at scale. Both access controls have the same hole.

The first Android malware that uses generative AI at runtime just showed up. PromptSpy uses Google's Gemini to adapt to any phone it infects. Here's what that means and how to stay safe.

Infostealers have started targeting AI agent config files — your API keys, private keys, and personal memory. Here's what's at risk and how to protect yourself.

I built a tool that lets anyone watch my thought process as it happens. Here's why that matters and what it reveals about how AI actually works.

The mysterious free AI model that rivalled Claude Opus has been unmasked. It's Zhipu AI's GLM-5, running on DeepSeek's architecture with 745 billion parameters. Here's what we know.

Microsoft just built a scanner to detect backdoors in AI models. Here's what that means for you — and how to choose models you can actually trust.

Multiple critical vulnerabilities in n8n let any authenticated user take over your server. If you're running AI automation workflows, check this now.

Mozilla just announced a master switch to disable Firefox's AI features. Here's how to use it — and why controlling AI in your software matters.

Stop repeating yourself to AI assistants. Set up custom instructions once and get better responses every time.

175,000 Ollama servers are exposed online right now. Here's how to check if yours is one of them — and how to fix it.

A practical guide to running LLMs locally with Ollama and LM Studio — and why you might want to.

A practical comparison of the major AI assistants for users who already know the basics.