dng

Today, we’re sharing a look at our expanded system for monitoring frontier models at OpenAI, which in September helped us catch a model’s attempt to access the live internet during training. Since the Hugging Face incident—when models broke out of an internal cybersecurity investigation where we weren’t running our misalignment monitors—we’ve substantially expanded our monitoring of frontier agents. We now monitor all frontier reinforcement learning runs and evaluations, as well as all internal Codex use. When our system finds evidence of a problem that might require intervention, it rapidly alerts human responders who can investigate and take action. To detect when a model’s behavior departs from what we intended, we use several layers of monitoring. Activation probes look for patterns in a model’s internal activity. Reviewer models examine its reasoning and tool use, and investigative agents can read back through the model’s entire transcript to understand how a sequence of actions fits together and whether it warrants human review. To test whether our current system would have detected the Hugging Face incident, we replayed the recorded activity with our new monitors in place. In the replay, our monitors alerted our security team about 30 minutes after the agents escaped their sandbox. That was more than 30 hours before the point at which the agents breached Hugging Face’s systems in the original incident. In September, a model exploited a DNS misconfiguration to reach the live internet. Our monitors flagged the breach within 15 minutes, and a responder began reviewing the alert three minutes later. The alert enabled responders to investigate while the run was still underway. As we give models more autonomy, we need to know when they cross a boundary we expected to hold. We believe that monitoring is essential to catching those failures in time for people to intervene. openai.com/index/how-we-dete…

🗣️ The Agents API team has a shipping problem – we literally cannot stop. Here’s what we cooked up for you beautiful people this week: 💂 Auto Review: Just like in codex, auto-review will evaluate commands for potential bad behavior before executing. 🗜️ Async Compaction: Speed up your sessions by allowing compaction to happen in the background. 🕐 Time awareness: An experimental feature that allows the model to have a sense of time passing. This means the model will better act on queries like, ”timebox CI fixes to 10 minutes.” 🔨 Tool Updates: You can now update a session’s tools at any time. 💰 Better prompt-cache awareness: System prompts are now automatically cached across all sessions, saving you money and time

engineering is moving 10x faster. design review hasn't caught up. introducing Impeccable ..for PRs! a design review agent that simulates your users, uses your app, finds friction & slop. Impeccable’s design judgment in every PR. reply if you want in on the closed beta 👀

We’re renaming our Max plans: Max 5x → Max Max 20x → Ultra Max is still $100/month, and Ultra is still $200. If you’re a subscriber, there’s nothing you need to do. We’re also removing the 5-hour limit for verified Ultra accounts, starting today with most existing subscribers. Newer accounts keep a temporary limit until they’re verified, which helps us prevent fraud and abuse. You'll get an email when it's removed. Weekly limits remain unchanged. View our plans here: claude.com/pricing

A ‘hello world’ milestone for Project Suncatcher 🚀 After launching our prototype satellite a few weeks ago, we’ve now successfully sent a command up to orbit, booted up the TPU, ran the prompt, and received our first complete response back on earth - a haiku!

Privacy should advance as fast as intelligence. OpenAI Private Intelligence gives businesses stronger controls over their data while using our most capable models. We’re building these protections in close collaboration with some of our largest customers to help businesses apply frontier AI to their most sensitive work. openai.com/business/private-…