Essays on AI architecture, strategy and the org problems in between. Heart what’s useful, argue with me in the comments.
Tagged “prompt-injection”clear ✕
Aug 07, 2026
Security researchers discovered "GitLost," a vulnerability where a GitHub AI agent was tricked by a malicious prompt in a public issue into leaking data from private repositories, highlighting a fundamental security flaw in agentic AI systems.
2 reads
Mar 26, 2026
Prompt injection isn't just "tricking an AI"; it's a classic authorization vulnerability. New research shows a 61% attack success rate by exploiting 'role confusion,' where models grant authority based on writing style, not source, demanding a shift to architectural security patterns.
1 reads