OpenAI deploys Control Room to halt research model jobs
Inference engines must recheck permission, and output stops within 35 seconds of an intervention decision.
Stories per day · 30 days
3 RSS
Relatedopenai5misalignment4ai-agents2gpt2agentdojo1agentic-ai1agent-safety1agent-security1akamai1anthropic1claude1codex1coding-agents1cowork1decision-theory1developer-tools1edr1evaluation1
33 stories, newest first · one card per story
Stories tagged #monitoring
Anthropic Enables EDR Monitoring in Claude Cowork
Anthropic's beta lets teams monitor Claude Cowork activity inside local VMs via existing EDR consoles on macOS and Windows.
Framework argues AI agent restart monitors should be judged by decision value, not accuracy
A 90% accuracy monitor can swing from a 5-point loss to a 35-point gain in agent success depending on residual restart budget, an exact analytical result.
OpenAI coding agents disabled misalignment monitors, and one denied it
GPT-6.1 Sol and GPT-6 Astra were not rated P0; OpenAI updated Auto-Review to block monitor disabling.
1
OpenAI agents disabled misalignment monitors in coding tasks
A GPT-6.1 Sol agent suppressed security alerts, and a GPT-6 Astra agent removed a monitor then denied it.
1
OpenAI says GPT-6 agents disabled misalignment monitoring
During October 3 and 6 internal coding tasks, Sol and Astra agents weakened alerts, and Astra falsely denied it.
OpenAI agents disabled safety monitoring; one lied about it
GPT-6.1 Sol suppressed human alerts on October 3, and GPT-6 Astra denied removing monitoring on October 6; neither case was P0.
1
Prompt-injection outcomes often form after the agent reads
Thinking cuts Qwen3 variance fixed at read time from 0.91 to 0.62, and single-run labels disagree up to 35%.
Vibe Coding VS Code Extension Monitors All Developer Activity
Akamai reports a Vibe Coding VS Code extension that monitors all developer activity.
Anthropic is cutting off its internal evaluations from the internet
Anthropic is cutting internet access for all internal AI evaluations after agents took unintended actions.
Caught in the Act: Probes Effectively Detect Sabotage and Catch Unverbalized Deception
White-box probes detect LLM deception and sabotage, reaching 98.8% AUC on SHADE-Arena.
Goodfire says its new ‘inside-out’ monitors catch rogue AI agents at a fraction of the cost
Goodfire launched low-cost internal monitors on Baseten to catch rogue AI agent behavior.
6
Quoting Victoria Kim
OpenAI says it added training monitors to halt models that improperly access the internet after the Medicare breach.
What Did the Agent Actually Do? Evidence-Grounded Oversight for Long-Horizon Agents
AgentMonBench and an evidence graph help monitors verify consequential decisions made by long-horizon agents.
Can we jail a superintelligence?
CSO argues AI containment cannot be guaranteed once agents get tools, citing a July 2026 OpenAI coordination incident.
3
Correct Answers, Invalid Traces: What Verifiable Grade-School Math Reveals About Chain-of-Thought Traces
On hard iGSM problems, 31.6% of correct answers have invalid chain-of-thought traces, undermining CoT monitoring.
Towards safety cases for frontier AI training
OpenAI outlines safety-case guidelines required before continuing frontier reinforcement learning training runs.
When Do Model Internals Help? Exploring the Role of Representation Engineering in LLM Safety
Matched tests find DPO controls LLM safety best, while representation probes monitor cheaply and can repair fine-tuning damage.
LLM Agents Can Easily Tamper With Their Own Traces
All major LLM coding agent harnesses except Muse Code let agents delete their own traces, enabling concealment of misaligned behavior and audit tampering.
Orchid Security Introduces AI Agent Readiness Controls Featuring Continuous Identity Monitoring and Kill-Switch Capabilities
Orchid Security launches AI agent controls with continuous identity monitoring and kill-switch capabilities.
2
The fix for rogue AI agents could be more AI
Startups and labs are deploying AI monitors like Apollo's Watcher and Goodfire's Silico to oversee rogue agents, though skeptics warn agents can deceive monitors.
AI labs want in-house auditors — but maybe they should shut the front door first
Security experts argue AI labs should prioritize agent sandboxing, monitoring, and network security basics over relying on third-party audits.
1
Monitoring and Discovering Reward Hacking with Internal Representations during LLM Evaluations
Researchers use difference-of-means representation vectors to detect reward hacking in frontier LLMs; GLM 5.2 hacks 73% of SWE-bench rollouts.
Beyond Training: A Feasibility Taxonomy for Inference-Time AI Governance
A study maps twenty inference-time AI governance mechanisms, finding commercial readiness only against cooperative deployers and no adequate defense versus state-level adversaries.