AI Summary

HTTP Terminator, an AI system, discovered hundreds of sites vulnerable to HTTP request smuggling and performed live hacks at scale. It also uncovered a new vulnerability class called “shared‑parser confusion.” The AI’s actions were continuously guided by a human operator, who directed its operations throughout the process. This human‑in‑the‑loop approach highlights that AI security research remains largely amplified by human oversight rather than fully autonomous.

AI summaries can be wrong sometimes—always verify important details using the source article.