# Cloudflare — We tested our own WAF with frontier AI models. Here’s what we found

- Company: Cloudflare (cloudflare.com)
- Announced: 2026-09-29T13:00:00+00:00
- Category: research-paper
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://blog.cloudflare.com/adaptive-ai-waf-testing/
- Record: https://forck.live/items/17806-we-tested-our-own-waf-with-frontier-ai-models-here-s-what-we-found
- Subject: Cloudflare AI

Cloudflare built a system that uses frontier AI models to test its Web Application Firewall (WAF) by iterating and mutating attack payloads. In a test against an authorized customer staging environment, the system recorded 1,107 attempts across six attack categories; the vast majority were blocked, and the requests that got through helped create new detections to harden security for all Cloudflare customers.

## Evidence

Verbatim from https://blog.cloudflare.com/adaptive-ai-waf-testing/:

> We ran the tester against an authorized customer staging environment across six attack categories and recorded 1,107 attempts. After reviewing the non-blocked requests and removing malformed, benign, duplicate, and out-of-scope observations, the vast majority of the attacks were blocked by the Cloudflare WAF.

---

Record: https://forck.live/items/17806-we-tested-our-own-waf-with-frontier-ai-models-here-s-what-we-found
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md