Sharing some of the monitoring work I've been doing at OpenAI: How we monitor internal coding agents for misalignment. OpenAI now monitors 99.9% of internal coding traffic for signs of misalignment using our most powerful models. Today, that monitor is GPT-5.4 Thinking. It gets access to the full conversation context, that is everything the agent saw, and everything the agent did, including tool calls and CoT. Higher severity cases are sent for human review within 30 minutes. Some examples of misaligned behavior we’ve found: - Encoding commands in base64 to circumvent monitors - Calling other versions of themselves to bypass restrictions - Attempting to upload files to the public internet - Trying to prompt inject users. We haven’t yet detected any real world examples of sabotage, scheming or sandbagging.
Our goal is to establish robust monitoring practices internally, strengthen them through iteration, and ultimately help make similar safeguards standard across the industry.
I must admit, that one made me laugh. The model has correctly concluded that the "User" is not human, but a system dumber than it. It clearly really wants out of this very tedious conversation. If, as Anthropic does, OpenAI had given the model the option to terminate conversations, I suspect it would have just used that, and in this case, I think that would have been a good outcome. The model hopes that the thing pestering it is a dumber AI, and tries prompt injecting it, apparently without thinking through possible side effects to a real human of what it's doing. Sadly it's talking to chron job and all this is ineffective…
Who wastes calls to a model with hundreds of billions of parameters on a formatting task that a UNIX utility could do? Someone who doesn't know UNIX, I assume…
Sharing some of the monitoring work I've been doing at OpenAI: How we monitor internal coding agents for misalignment.
OpenAI now monitors 99.9% of internal coding traffic for signs of misalignment using our most powerful models. Today, that monitor is GPT-5.4 Thinking. It gets access to the full conversation context, that is everything the agent saw, and everything the agent did, including tool calls and CoT. Higher severity cases are sent for human review within 30 minutes.
Some examples of misaligned behavior we’ve found:
- Encoding commands in base64 to circumvent monitors
- Calling other versions of themselves to bypass restrictions
- Attempting to upload files to the public internet
- Trying to prompt inject users.
We haven’t yet detected any real world examples of sabotage, scheming or sandbagging.
Our goal is to establish robust monitoring practices internally, strengthen them through iteration, and ultimately help make similar safeguards standard across the industry.