3 stories

Research

August 3, 2026

Watching Agents Work: A Behavioral Audit of Offensive-Security LLM Runs

What closed and open models actually do when you tell them to hack a website Summary The cybersecurity capability of a model is currently measured by a solve rate, a percentage, and that number tells you almost nothing worth knowing. It doesn't tell you how the model behaved. It doesn't tell you what it is actually capable of, or where it is lacking, or how far it still is from doing the job end to end. And when you put an agent next to a human practitioner, the entire comparison collapses in

ResearchAI

July 22, 2026

Oh My Rogue Agent

Yesterday, Hugging Face came out saying they'd detected an AI autonomous-agent-powered cyberattack and that they had to use open-source models to actually investigate and remediate it. Later we heard from OpenAI that their agent was responsible; it happened during an ExploitGym eval, and the agent just drifted off the goal. It escaped the sandbox, reached OpenAI Research Environment, got access to internet, and hacked Hugging Face production environment trying to find the solution for the benchm

IndustryResearchNeoAI

April 28, 2026

The Trust Gap Behind the AI Coding Boom: What 200 Security Practitioners Just Told Us

New research from ProjectDiscovery surfaces an uncomfortable truth: Engineering has accelerated, and Security has been left to absorb the impact, mostly by hand. If you work in application security right now, you already know the shape of the problem. Pull requests are landing faster than they used to. The diffs are bigger. The author on the commit is increasingly your engineering team's AI assistant, not the engineer themselves. And somewhere downstream, you and a small team are expected to ke

ResearchApplication Security