nFOX Lab open-sources AI security sandbox Blockhouse

AI governance startup nFOX Lab has launched a platform for managing enterprise large language models and announced plans to open-source Blockhouse, a security sandbox that restricts what AI workloads can access. Blockhouse operates at the Linux kernel level, removing privileges by default and requiring administrators to restore access deliberately. Communications pass through a monitored channel called Porthole, and model files are sealed with SHA-256 hashes to verify integrity. nFOX will release Blockhouse under the Apache 2.0 license while charging for its fleet management platform. The startup is self-funded with paying beta customers but is not yet profitable.