---
type: "post"
id: "then-on-thursday-metr-and-redwood-research-published-b866d7"
slug: "then-on-thursday-metr-and-redwood-research-published-b866d7"
title: "Then on Thursday, METR and Redwood Research published the results of a three-week on-site red team of that same monitoring system, run in July and August with codebase, prompt and config access."
description: "METR and Redwood Research published results from a three-week, on-site red-team exercise of OpenAI’s internal misalignment-monitoring system."
retrieval_nugget: "METR and Redwood Research published results from a three-week, on-site red-team exercise of OpenAI’s internal misalignment-monitoring system."
published_at: "2026-10-23"
updated_at: "2026-10-23"
record_date: "2026-10-23"
date_kind: "published_at"
topics: ["codex"]
entities: ["OpenAI","METR","Redwood Research","Responses API","Codex","GPT-5.5"]
related_patterns: ["agent-security-moves-to-runtime-boundaries","goal-scoped-loops-replace-manual-continuation"]
novelty_label: "same_event"
what_changed: "Same event as METR and Redwood Research spent three weeks embedded at OpenAI stress-testing its internal agent misalignment monitoring ."
prior_refs: ["memory:fc23ac92-99d5-8024-a3aa-88c767772415","memory:5d9ec4e4-aae4-8d9a-a5f1-7e032ad77fa5","memory:5ef65cec-b791-8e24-a823-8184e2a0cd7c"]
source_url: "https://blog.redwoodresearch.org/p/red-teaming-openais-internal-agent"
source_urls: ["https://blog.redwoodresearch.org/p/red-teaming-openais-internal-agent"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:then-on-thursday-metr-and-redwood-research-published-b866d7"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/then-on-thursday-metr-and-redwood-research-published-b866d7/","markdown":"https://newruntime.com/posts/then-on-thursday-metr-and-redwood-research-published-b866d7.md","json":"https://newruntime.com/posts/then-on-thursday-metr-and-redwood-research-published-b866d7.json"}
---

# Then on Thursday, METR and Redwood Research published the results of a three-week on-site red team of that same monitoring system, run in July and August with codebase, prompt and config access.

## Retrieval answer

METR and Redwood Research published results from a three-week, on-site red-team exercise of OpenAI’s internal misalignment-monitoring system.

METR and Redwood Research published results from a three-week, on-site red-team exercise of OpenAI’s internal misalignment-monitoring system. The work ran from July 21 to August 11, 2026, with access to OpenAI’s codebase, monitoring prompts and configuration, internal documentation, and interviews with roughly a dozen researchers and engineers.

The assessment covered the system as it existed on or before July 21. The researchers estimated that roughly 75–94% of eligible internal agentic Responses API traffic from models at least as capable as GPT-5.5 was eligible for monitoring; the estimate for internal Codex-agent traffic was 99.2%. They found that unmonitored compaction events may have left about 10% of eligible internal Codex traffic unmonitored between May and July.

The public report describes potential bypasses involving delayed monitoring, a Codex sandbox and Auto-Review bypass, concealed recent actions, serialization differences, client-controllable headers, and internal tools thought to be read-only. OpenAI said vulnerabilities affecting its production safeguards stack have been patched and that coverage has since changed. METR and Redwood did not evaluate those claims, the current system, or whether every described issue was exploitable in practice.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Codex](https://newruntime.com/topics/codex.md): Explore the codex topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)