AI-Powered Attacks Go Autonomous: Start With Credentials
Fully autonomous cyberattacks may be only months away. The incidents happening right now show exactly where defenders should focus their attention.
Fully autonomous cyberattacks may be only months away. The incidents happening right now show exactly where defenders should focus their attention.
The FBI’s ARTEX alert shows AI agents reaching privileged access in under three hours, while leaked credentials stay valid for years. Here’s what that gap means for defenders and how to close it.
AI coding agents can read files, call tools, and use credentials that never appear in a pull request. For AppSec teams, securing the code is only part of the job.
Learn how AntiFox and ARTEX use agentic AI to find and exploit exposed credentials, what AWS intrusions reveal, and how security teams can reduce credential exposure.
An ARTEX build ships no secret scanner. In our lab, it found 8 of 9 planted secrets in seconds, even when asked only to assess the repo.
~45 agentic attack frameworks beyond ARTEX, fingerprinted for internet scanning: dozens of operator consoles sit exposed today, many unauthenticated.
31,088 secrets in Claude Code’s .claude/settings.local.json on public GitHub, 5,343 valid at detection, 23,276 repositories. 613 more in Docker Hub images, 99 valid. Validity, timeline, Docker Hub, who is affected.
Microsoft Threat Intelligence caught attackers using an agentic AI framework, ARTEX, to hunt exposed secrets on public GitHub and move into Azure tenants in as little as five minutes. GitGuardian has tracked this attack path since 2017. Here is what changed, and four things to do today.
GitHub’s Octoverse 2026 counted 34.3 billion commits, and over 40% of pull requests now start with an AI agent. Read to learn what the State of the Octoverse 2026 Report reveals about AI, citizen developers, and our new security reality.
Researchers pulled real, leaked API keys out of three LLMs using only API output. Here is how AI agents carry secrets from developer laptops into model context, and how to stop them before the model sees them.
Six AI agent security best practices for teams running coding agents and MCP servers: limit what a compromised agent can reach and catch credential misuse.
An AI agent runs with your credentials and acts before you can stop it. Nine incidents show what that costs. Six controls limit the damage: sandbox the agent, scope its credentials, lock its configuration, log every call, scan the workspace first, and deny by default.