Show HN: Flight Risk: Can you break an AI agent?

Category: ai-ml

Tags: ai-security, capture-the-flag, adversarial-testing, interactive-demo

Score: 6.0/10 (Innovation: 7, Technical: 6, Documentation: 3, Utility: 5)

Flight Risk is an interactive CTF-style challenge where users attempt to break or jailbreak an AI agent that gets progressively smarter across six rounds. It's interesting because it gamifies AI safety testing and provides hands-on experience with adversarial prompting techniques.

Target audience: ai-researchers, security-engineers, ml-engineers, developers-testing-ai-systems

Repository: https://ctf.demo.lorikeetcx.ai/

View on Hacker News