Cloudflare is wiring AI directly into internet-scale security. The company, which proxies roughly 20% of the web’s traffic, announced that it is combining OpenAI models accessed through Daybreak with its global network and threat intelligence to find vulnerabilities and propose fixes — and that its engineers are using Codex to build faster.
Cloudflare security (Credits: OpenAI)
From finding bugs to fixing them
The centerpiece is Vulnerability Discovery and Remediation, an early-access service offered through Cloudflare Managed Defense. Here’s how it works: the OpenAI models investigate customer-authorized code and assess potential vulnerabilities, while Cloudflare layers in live traffic data, security signals and validation to prioritize findings. The system then prepares proposed code patches — and, where appropriate, custom web application firewall rules to reduce exposure — for customers to review.
Crucially, customers keep control: Cloudflare’s team reviews the proposals before sharing them, and it is up to the customer which changes to implement. The goal is a clear path from identifying a vulnerability to protecting the affected application.
“Machines do the grunt work”
Cloudflare is also betting big on AI inside its own engineering org. According to Mark Anderson, Cloudflare’s President of Revenue, a third of the company are software developers — and 100% of them use Codex.
The payoff can be dramatic. Anderson described a product rebuild the team had postponed for lack of manpower: with Codex, four developers built it in three days, spending just $1,100 on tokens — work Anderson estimated would otherwise have required months of sequencing and “maybe 10x more developers.”
“The beauty of the automation that Daybreak and Codex provide is that you can assign machines to do the grunt work so humans can exercise their judgment on the most important engineering and security decisions,” Anderson said.
Also Read: OpenAI Wants One Set of Safety Rules for Every Platform Distributing Its AI
Source: OpenAI — read the full story there.
