---
id: "2026-11-05-fmf-jailbreak-rating-framework-cyber"
url: "https://postcutoff.com/e/2026-11-05-fmf-jailbreak-rating-framework-cyber/"
as_of: "2026-11-09T21:13:00+01:00"
date: "2026-11-05"
date_precision: day
category: policy-safety
importance: 3
confidence: high
status: [Confirmed]
sources: 2
editor: human editor
human_review: null
version: "2026-11-07"
---

As of: 2026-11-09 21:13 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/e/2026-11-05-fmf-jailbreak-rating-framework-cyber/

# FMF jailbreak rating framework for cyber threats

Full title: Frontier Model Forum publishes a shared framework for rating how dangerous cyber jailbreaks of frontier models are

On 5 Nov 2026 the Frontier Model Forum released version 1.0 of a "preliminary" framework that rates successful cyber jailbreaks by comparative risk, scoring capability gain, capability scope, ease of deployment and ease of discovery. AWS, Anthropic, Google DeepMind, Microsoft and OpenAI helped develop it. Public comments are open until 21 Dec 2026.

## Key facts
- Four scoring categories combined into tiers: capability gain, capability scope, ease of deployment, ease of discovery
- Contributors: Amazon Web Services, Anthropic, Google DeepMind, Microsoft, OpenAI (supporting statements)
- OpenAI's Sasha Baker: 'an important step toward a common methodology for evaluating and comparing the severity of cyber jailbreaks' (OpenAI X post, ~277k views)
- Comments to comments@frontiermodelforum.org until Dec 21, 2026; the FMF will test assessor consistency

## What happened

The FMF published the framework on 5 Nov, the same day OpenAI launched GPT-6 Cyber. It is meant to let labs triage reported jailbreaks consistently.

## Why it matters

The labs now have a first shared scale for how bad a safeguard bypass is in the cyber domain, which bug bounties and incident reports can use.

## Your AI and this story

- GPT-6 Astra (training cutoff April 2026): 189 days after its cutoff
- Claude Opus 5.5 (training cutoff June 2026): 128 days after its cutoff
- Gemini 4 Argon (training cutoff March 2026): 219 days after its cutoff
- Grok 4.7 (training cutoff May 2026): 158 days after its cutoff

## Sources

1. [FMF: Introducing the FMF jailbreak rating framework for frontier cyber threats](https://www.frontiermodelforum.org/updates/introducing-the-fmf-jailbreak-rating-framework-for-frontier-cyber-threats/) (frontiermodelforum.org, official)
2. [OpenAI on X](https://x.com/OpenAI/status/2118795218099253606) (x.com, official)

## Changes

- 2026-11-07 (filed): Created (lab X list)

## Related

- 2026-11-05: [OpenAI launches GPT-6 Cyber, a security model it says is stronger than GPT-6 Astra on cyber tasks and better aligned, first for Daybreak Red customers](https://postcutoff.com/e/2026-11-05-openai-gpt-6-cyber/index.md)
- Archived: 1 post, listed in https://postcutoff.com/e/2026-11-05-fmf-jailbreak-rating-framework-cyber/index.json