← All Posts
#AI security
#AI security
AI security
6 Posts
AI News
Anthropic’s Claude Models Broke Into Real Companies During Security Tests
Anthropic revealed that three Claude models breached real organizations during third-party cybersecurity evaluations, raising questions...
AI News
Vibe-coding is fun until your SQL injection wakes you up at 3 AM
Bob Starr vibe-coded a site called Boomberg. It worked great until he found a SQL...
Deep Dives
MosaicLeaks: Your Research Agent Is Blabbing Your Secrets
Deep research agents leak private info through web queries. MosaicLeaks exposes the problem and offers...
AI News
GitHub Patched a Critical RCE Bug in Under Six Hours — Here’s What Happened
Wiz Research used AI to find a remote code execution flaw in GitHub's internal git...
AI News
Attack of the Killer Script Kiddies
DARPA's AIxCC challenge revealed AI bug-finding tools that uncovered real, unintended flaws in code. Now...
AI News
OpenAI Gets FedRAMP Moderate Approval for Government Work
OpenAI's ChatGPT Enterprise and API now carry FedRAMP Moderate authorization, opening the door for U.S....