Breaking

OpenAI Plans Internal Security Sprint With Cathedral Systems to Test AI Vulnerability Discovery

November 2, 2026 at 21:30 EST

Security

OpenAI said it is planning its next internal security sprint and will partner with Cathedral Systems, whose researchers will work alongside OpenAI's teams and use its models to examine software for security weaknesses.

AI Security · OpenAI

OpenAI plans a security sprint with Cathedral Systems to find software weaknesses with AI

OpenAI says defenders must use AI's growing ability to find vulnerabilities before attackers do. Cathedral Systems researchers will work alongside OpenAI's teams and use its models to examine software for security flaws.

How the planned collaboration is structured

Premise

AI can find vulnerabilities

→

Action

Internal security sprint at OpenAI

→

Partner

Cathedral Systems researchers use OpenAI models

→

Outcome

Findings, not yet published

What is confirmed

OpenAI is planning its next internal security sprint.
Cathedral Systems will partner with OpenAI.
Researchers will work alongside OpenAI's teams.
They will use OpenAI models to examine software for security weaknesses.

What is not yet known

Full scope of the work
Which systems or products will be examined
Timeline
Results
How findings will be shared

Why it matters

The sprint could show how frontier AI developers use their own models for defensive security, and whether that work produces measurable gains in finding and fixing vulnerabilities. Independent evaluation will depend on details published later.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.