---
title: "What "Private AI" Actually Means: The University of Florida's NaviGator in Practice"
description: "5 source-backed AI claims from What "Private AI" Actually Means: The University of Florida's NaviGator in Practice, with key statistics, context, and the…"
url: "https://kbanc.com/claims-library/private-ai-university-of-florida-navigator"
source: "https://aiadopters.club/p/how-a-university-made-ai-safe"
date: "2026-07-25"
topics: ["strategy", "implementation", "business"]
generated: "2026-09-22"
---

# What "Private AI" Actually Means: The University of Florida's NaviGator in Practice

By Kamil Banc | July 25, 2026

## Claims

1. **Data Tagging Protects Access**: UF tags all 104 AI models with specific permitted data access rules for protection.
2. **NaviGator AI On HiPerGator**: The University of Florida built NaviGator AI on its HiPerGator supercomputer for compliance.
3. **Massive Hardware Investment**: UF invested $70 million in 2020 and $33 million later on Blackwell hardware.
4. **Governance Over Hardware**: Data tagging, not expensive hardware, is what actually protects sensitive information at scale.
5. **Preventing Policy Workarounds**: Easy-to-use approved AI workflows prevent employees from bypassing security policies with unauthorized tools.

## Evidence

### Quote
> "But the supercomputer isn’t the part protecting the data." - Kamil Banc

### Key Statistics
- **104 AI models**: Number of AI models UF tags with specific data access permissions for governance.
- **$70 million**: Initial AI initiative investment by the University of Florida in 2020.
- **$33 million**: Additional spending on Blackwell hardware to expand AI infrastructure.
- **$6 million/year**: Approximate annual cost to cool and operate the HiPerGator supercomputer.

## Context
The case study examines how the University of Florida manages sensitive data—including FERPA-protected student records, clinical data, and export-controlled research—within its AI ecosystem, NaviGator AI. Rather than relying solely on the raw computing power of its HiPerGator supercomputer, UF applies granular data-tagging policies across all 104 AI models to control what information each model can access. This approach demonstrates that governance frameworks, not just infrastructure investment, are the critical mechanism for securing enterprise AI deployments. Practitioners can apply this model by prioritizing metadata tagging and access control policies before or alongside hardware investment, ensuring usability so employees don't bypass official tools.

## Source
- Original: [What "Private AI" Actually Means: The University of Florida's NaviGator in Practice](https://aiadopters.club/p/how-a-university-made-ai-safe)
- Cite: kbanc.com/claims-library/private-ai-university-of-florida-navigator