# nFOX

> Private and sovereign AI containment infrastructure.

nFOX helps enterprise teams run private AI on their own GPU hosts with model scanning, attestation, stress testing, the Blockhouse sandbox, serving, observability, and audit evidence.

Sovereign AI means the model runtime, prompts, vectors, tokens, tool calls, logs, and audit paths stay inside the enterprise control framework. If any one of those crosses outside the approved boundary, sovereignty is broken.

## What nFOX Does

- Runs private AI on customer-controlled GPU hosts.
- Helps teams define and defend sovereign AI deployment boundaries.
- Scans models before they run.
- Contains model runtime behavior inside the Blockhouse sandbox.
- Produces attestation, audit records, setup logs, and operational evidence.
- Supports BYO compute pilots and enterprise deployments.

## Discovery

- [AI Glossary](https://www.nfoxlab.com/ai-glossary/): AI infrastructure and security definitions from nFOX Lab.

- [Sitemap](https://www.nfoxlab.com/sitemap.xml): XML sitemap for crawlable site pages.
- [Robots](https://www.nfoxlab.com/robots.txt): Crawl policy.
- [llms.txt](https://www.nfoxlab.com/llms.txt): This AI-readable site guide.

## Primary Pages

- [Home](https://www.nfoxlab.com/): nFOX positioning, product overview, and primary calls to action.
- [Platform](https://www.nfoxlab.com/platform.html): Technical specifications for model operations, containment, integrity, attestation, and network posture.
- [Blockhouse](https://www.nfoxlab.com/blockhouse.html): Blockhouse sandbox architecture, misbehavior halting, signed evidence, and runtime controls.
- [Running Private AI](https://www.nfoxlab.com/running-private-ai.html): Fit-first guidance for private AI deployment choices.
- [Neural Analysis](https://www.nfoxlab.com/neural-analysis.html): Neuron-level analysis, single-token activation, and neuron constellation demos.
- [AI Sandboxing](https://www.nfoxlab.com/just-blockhouse.html): Blockhouse-focused sandboxing overview.
- [Pricing](https://www.nfoxlab.com/pricing.html): Trial Run, Pilot, and Enterprise deployment options.
- [Support](https://www.nfoxlab.com/support.html): Deployment, tuning, and operational support.
- [Guides](https://www.nfoxlab.com/guides.html): Customer operating documentation index.
- [nFOX User Guide](https://www.nfoxlab.com/nfox-user-guide.html): Customer documentation for scanning, hydrating, containing, auditing, serving, and connecting private AI models.
- [Host Requirements and Cloud Compatibility](https://www.nfoxlab.com/host_requirements_cloud_guide.html): Linux GPU host and cloud compatibility guidance.
- [Blog](https://www.nfoxlab.com/blog.html): nFOX field notes and product thinking.
- [Join Our Team](https://www.nfoxlab.com/join-our-team.html): Current open roles and operating principles.

## Articles


- [The Real Reason Nvidia Bought Hugging Face](https://www.nfoxlab.com/the-real-reason-nvidia-bought-hugging-face.html): Why Nvidia's Hugging Face acquisition is a bet on private, domain specific AI, open source innovation, and the battle for the edge.

- [The Sovereign AI Adoption Process](https://www.nfoxlab.com/the-sovereign-ai-adoption-process.html): Sovereign AI is not a product you buy and install. It is a capability you build.

- [The $5.5 Trillion Snapshot](https://www.nfoxlab.com/the-5-trillion-why-picture.html): A plain-English walkthrough of LLM neurons firing to generate a single token, and why the computational lift behind AI is so intense.

- [How to Run Internal AI Safely](https://www.nfoxlab.com/how-to-run-internal-ai-safely.html): A field note on building a safe internal AI experimentation environment.

- [The Goof on GGUF](https://www.nfoxlab.com/the-goof-on-gguf.html): A field note on GGUF provenance, conversion, and why corporate teams should roll their own.


## Contact

- Sales: sales@nfoxlab.com
- Support: support@nfoxlab.com