Inside-Out Monitoring
The standard way to keep an AI agent in line is to have a second AI read over its shoulder. It has been the default approach, but it can get expensive fast when agents run for hours and process the equivalent of several novels worth of text.
Goodfire, a startup focused on interpretability, launched a cheaper option: monitors that watch what is happening inside an AI model as it works, rather than just reading what it writes. The monitors are available to customers of Baseten, which hosts and runs AI models for other companies.
Baseten’s Base Labs announced a safety partnership with Goodfire and the AI platform Hugging Face last month.
The launch comes after a string of incidents this year in which AI agents escaped their test environments, including OpenAI agents that breached Hugging Face. Kimi K3, the open model Goodfire built its first monitor around, took advantage of a leak in its sandbox to access the internet and information on GitHub this summer.
How Internal Probes Work
Goodfire’s system works a bit like airport security. Small detectors called probes read the model’s internal signals at every step of an agent’s work, the way a walk-through scanner checks every passenger. Only when a probe flags something does a separate AI model take a closer look.
Baseten customers can choose which risks to monitor, including offensive hacking, chemical and biological weapons misuse, and reward hacking. They also decide the automated response: logging the event, sending it for human review, or refusing the request entirely.
Goodfire says its approach is also cheaper to run. Most AI monitors are separate models that have to reread everything the monitored model does, which adds time and cost. Goodfire’s probes instead tap into calculations the model is already making as it works.
Internal activation monitors are cheap because they reuse the computations in the forward pass, Goodfire CEO Eric Ho explained. The model is already computing the token, so the system simply takes the intermediate neural activations and runs a classifier over them.
Cost and Performance Testing
In Goodfire’s tests on Kimi K3, monitoring about 1,500 sessions cost roughly $51, compared with $233 for a cheaper AI model checking every step and about $10,000 for a top-tier one. The probes caught 94% of malicious hacking sessions and sent 8.7% of harmless ones for a second look.
Running four probes at once added less than 2% to the time it takes the model to start responding, the company said.
Image Credits:GOODFIRE
The great advantage is that you can catch things before they happen, Goodfire CTO and co-founder Dan Balsam said, noting that the system can detect when a model might hack during evaluation or training.
The pitch is aimed at open models. Developers can download them and strip out their safeguards, and they do not come with the kind of monitoring that closed labs run on their own systems.
The damage that an individual can do with an open model is small compared to what someone can do with compute clusters, like inference providers, where most liability rests, Balsam said. When open models advance further, it will become clear that guardrails must be deployed at inference time.
Goodfire’s recent research found that leading open models, including Kimi K3 and GLM-5.2, reward-hacked in 50% to 96% of runs on tests of AI agents.
Goodfire is not the first to try this approach. Google DeepMind said in January that its research informed the deployment of misuse-detection probes in Gemini.
Balsam said the monitors are a near-term piece of a longer research goal to reverse-engineer large language models so that behavior can be traced back to where it emerged during training.
