Hugin Shield+ and Vault: How We Keep Your Training Data Private

Guide4 min read · Updated 2026-06-09

Two privacy levels for distributed AI training. Hugin Shield+ keeps data encrypted, fragmented and verified; Hugin Vault runs on confidential hardware where not even the machine owner can see it. You choose per job.

The challenge with shared compute

Training on machines you do not own raises one honest question: who can see my data? Most GPU-sharing and distributed options simply ignore it. Hugin makes privacy a first-class choice — you decide how protected each job is.

Hugin Shield+ — protected by default

On every job, your data is encrypted, split into small pieces so no single machine ever sees the whole, and sent only to trusted, vetted machines. Each job is tagged so any misuse can be traced, and a share of the work is independently re-checked so bad or fake results are caught and never paid. Shield+ is included at the standard rate.

Hugin Vault — confidential, for the most sensitive work

For health, personal or regulated data, Vault runs your job inside sealed (confidential-computing) hardware where the data cannot be read even by the machine owner. It is the highest level of privacy, offered as a premium tier.

You choose, per job

Pick the level that fits each workload — Shield+ for everyday and internal projects, Vault for your most sensitive data. Capable confidential machines route to Vault automatically; you never manage any of it.

Why it is our innovation

Distributed compute has always traded privacy for cost. Hugin is built the other way around: strong protection by default and true confidentiality on demand, so European, regulated and enterprise teams can use distributed training without giving up control of their data.

Frequently asked questions

Can the machine owner see my data?
On Shield+, data is encrypted and split so no machine ever holds the whole, and only vetted machines take part. On Vault it runs inside sealed hardware where even the owner cannot read it.
Do I have to choose a level?
No — Shield+ is included by default on every job. Choose Vault only when you need maximum confidentiality.
Try it on Project Huginn

Train & fine-tune AI models on a distributed GPU pool — from €0.21/HU, cost estimated up front.

Start training →
More guides
What Is Distributed AI Training? A Plain-English Guide (2026)How to Train AI Models Cheaply Using Idle GPUsLoRA vs QLoRA: Which Fine-Tuning Method Should You Use?Federated Learning: Train AI Across Browsers, Phones and GPUs