Kimi K3 found Redis RCE and assembled a working exploit in 27 minutes

🛡 Kimi K3 found Redis RCE and assembled a working exploit in 27 minutes

A swarm of 32 AI agents based on the Kimi K3 model (2.8T parameters, MoE) from Moonshot AI discovered RCE vulnerabilities in Redis, including an exploit for version 8.8.0, assembled in just 27 minutes.

🌍 This demonstrates a sharp transition toward autonomous discovery and the creation of working exploits (offensive AI), which necessitates a revision of security strategies.

👤 AI agents are becoming a real threat in cybersecurity, capable of finding critical holes in popular services within minutes.

Source 1: https://thehackernews.com/2026/07/kimi-k3-agents-found-redis-zero-days.html Source 2: https://github.com/berabuddies/redis-poc