Research and white papers

Building AI you actually own.

Field notes, reference architectures and white papers on running AI you actually own, sovereign, on-premise and built for regulated industries.

ReportFeatured

Local AI in Sweden and the EU: The 2026 Guide to Sovereign Deployment

Why Swedish and European organisations are moving AI on-premise in 2026, the regulation driving it, who is deploying local AI, and how to choose a sovereign platform.

Sovereign AI4 July 20266 min read

Read the paper

12 published

Article

Sovereign AI: What It Means and Why Regulated Enterprises Need It

A plain-language primer on sovereign AI, what it is, how it differs from cloud AI, and why regulated organisations are bringing their models in-house.

Sovereign AI24 June 20263 min read

White Paper

On-Premise AI: A Reference Architecture for Air-Gapped LLM Deployment

A practical reference architecture for deploying large language models on-premise and air-gapped, network topology, GPU sizing, model serving and zero-egress design.

Architecture20 June 20263 min read

Report

Digital sovereignty in the EU: a 2026 field guide

Europe spent 2025 and 2026 waking up to its dependence on a few foreign cloud providers. A field guide to the laws, the money and the moves that follow.

Sovereign AI18 June 20263 min read

White Paper

The EU AI Act and Self-Hosted LLMs: A Compliance Checklist

How the EU AI Act maps onto self-hosted and on-premise LLM deployments, a practical, plain-language checklist for legal, compliance and security teams.

Compliance16 June 20263 min read

Article

Model Provenance and Supply-Chain Trust for Private LLMs

Why model provenance matters for private LLM deployments, and how signing, hashing, pinning and trust scoring keep the weights you reviewed the weights you run.

Security11 June 20262 min read

Report

The real cost of on-premise AI versus per-seat cloud assistants

A grounded look at when running open models on your own hardware beats paying per seat or per token, and when it does not, with the break-even maths.

Economics20 May 20264 min read

Report

Benchmarking open-weight models for regulated back-office work

Open models are good enough for a lot of real enterprise work. A practical guide to proving that for your own tasks instead of trusting a leaderboard.

Architecture13 May 20263 min read

White Paper

The security case for keeping AI inference on-premise

A threat model for enterprise AI: what actually goes wrong, why shadow AI is a breach vector, and how keeping inference in-house shrinks the attack surface.

Security29 April 20263 min read

White Paper

SOC 2 and ISO 27001 for a self-hosted AI platform: mapping the controls

What SOC 2 and ISO 27001 actually certify, how they differ, and how a self-hosted AI platform maps onto their controls instead of inheriting a vendor's.

Compliance8 April 20263 min read

White Paper

Data protection by design for LLM deployments: a technical white paper

How GDPR principles like minimisation, erasure and lawful basis actually land on a working AI system, and why architecture decides whether you can comply.

Compliance25 March 20263 min read

Report

Estimating your real LLM token bill: a working methodology

Most teams underestimate what cloud AI will cost at scale because they price the demo, not the workload. A method for estimating the bill before you sign.

Economics11 February 20263 min read

Research from the frontier of sovereign AI, in your inbox.

Occasional, substantive notes on on-premise AI, model provenance and EU compliance. No spam, and, fittingly, we host it ourselves.