Metacognition Releases Open Source ZEOS Runtime for AI Agents

Sep 30, 2026
ZEOS runs existing language models while enforcing policies on agent actions and communication. The open source runtime requires no model retraining or new hardware.

Metacognition released ZEOS under an open source license on September 30, the company said in a press release. The runtime works with existing large language models and serving systems, without requiring model retraining or new hardware.

ZEOS lets AI agents propose actions, then checks those proposals against configured policies before execution. Proposals outside the permitted boundaries are ignored, while agents can communicate only through channels declared in advance.

The runtime records rejected actions and their causes for auditing. Metacognition says ZEOS is not an AI agent or model and cannot be prompted to bypass its policy settings.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Cybersecurity AI Weekly or Daily AI Brief.

Also, consider following us on social media:

Free newsletter

Cybersecurity AI Weekly

Weekly newsletter about AI in Cybersecurity.

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more
Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.