Field notes from inside a sovereign AI.
Mickai® is a Sovereign Intelligence Operating System (SIOS) that runs on your own hardware. These are long-form essays on the architecture of the Mickai SIOS, the patterns that keep surfacing in commercial AI in 2026, and the engineering choices that make sovereign intelligence possible. Written by Micky Irons, named inventor of the 104 filed UK patent applications, recorded on the UK IPO public register at numbers GB2607309.8 to GB2611702.8, GB2611885.1 onwards, GB2612762.1 to GB2612793.6, GB2613386.8 to GB2613404.9, and GB2615041.7 to GB2615043.3.
The cooperative the field notes describe, running on the sovereign silicon substrate.
Showing 73 to 96 of 1124 articles.
The Honest Way to Prove an Audit Log Was Not Altered, Offline
Why we build tamper-evidence instead of promising the impossible, and how a signed off-box checkpoint turns a hash chain into proof you can trust without a network.
Why We Filed 104 Patent Applications on the Mechanism, Not the Model
The durable inventions in a sovereign AI system are not the models. They are the ways trust is enforced offline. That is what our filed estate protects.
How Airports Meet EASA Part-IS from February 2026 with On-Site AI
Airports meet EASA Part-IS by owning information-security risk themselves and running AI on airport hardware so no passenger or operational data leaves the boundary.
Can Automotive Suppliers Use AI on OEM Design Data While Keeping TISAX Prototype Protection?
Yes, provided the AI runs on the supplier's own hardware so OEM design and prototype data never leaves the TISAX protected zone.
AWS Bedrock versus sovereign on-premise AI: is 'we do not train on your data' enough for regulated workloads?
A no-training promise addresses one reuse path but leaves CLOUD Act reach and verifiable audit unsolved, which is why regulated workloads need sovereign control.
Should We Build Our Own On-Premise AI With Ollama Or Buy A Sovereign Operating System?
Run Ollama to prototype, but buy a sovereign operating system when the model must survive an audit and prove who did what.
Can councils run AI on resident and social care records without adding a cloud data processor?
Yes: councils can run AI over social care records on their own hardware, adding no cloud processor, because the records never leave the building.
Is CPU or GPU inference cheaper for on-premise enterprise AI?
For everyday mixed enterprise workloads CPU inference is usually cheaper per query, and a GPU only pays off at sustained high utilisation.
How To Meet the UK Data (Use and Access) Act 2025 Rules on Automated Decisions
The Act reforms Article 22 but keeps the safeguards, so a sealed on-premise audit ledger proves each automated decision's basis without exporting personal data.
Can energy operators run AI on grid and OT data on-premise to satisfy the Cyber Assessment Framework?
Yes: forecasting and anomaly detection that run entirely on operator-owned hardware keep grid and OT data inside the Cyber Assessment Framework boundary.
Can hedge funds run AI on proprietary models and trade data without leaking alpha to a cloud vendor?
Yes: run sovereign models on hardware the fund owns, and a hedge fund's alpha and MNPI never leave its control.
Is Microsoft 365 Copilot sovereign, or does regulated data still leave the building?
No. Copilot keeps most data in the EU, but the US CLOUD Act can still reach it, so regulated inference belongs on owned hardware.
What ISO/IEC 42001 Certification Requires, and Why On-Premise AI Makes It Easier to Pass
ISO/IEC 42001 requires an audited AI management system, and on-premise AI turns several Annex A controls into evidence generated automatically.
How Crypto Firms Meet MiCA Record-Keeping and Market-Abuse Duties Without Cloud AI
A crypto firm meets MiCA's record-keeping and market-abuse duties by running monitoring on its own hardware behind a zero-egress perimeter, exporting no transaction data.
What Does On-Premise Enterprise AI Actually Cost to Deploy in 2026?
On-premise enterprise AI splits into four costs: hardware, model licensing, integration, and staff, and the recurring integration and staff bills usually dwarf the hardware.
On-premise fraud detection, in motion: watch Nemesis catch a laundering ring before the money moves
A demo of the sovereign fraud and crime detection studio that scores every payment on your own hardware, explains itself, and seals a record a regulator can verify offline.
Can We Run Online Safety Act Moderation Without Sending Content to Cloud AI?
Online Safety Act duties can be met by on-device classifiers that keep user content and evidence inside your own jurisdiction.
How Payment Processors Keep AI Inside PCI DSS 4.0 Scope Without Exporting Cardholder Data
Run the model on operator-owned hardware inside the cardholder data environment, so cardholder data never leaves and the PCI DSS assessment stays small.
How to bring your AI models under the PRA's SS1/23 model risk rules
Register each AI model in one inventory, tier it, and validate it independently, which is straightforward when every model runs on hardware you own.
Is a private Azure OpenAI deployment actually sovereign?
A private Azure OpenAI deployment is isolated but not sovereign: the model still runs on vendor hardware under US jurisdiction with vendor-held keys.
How Much GPU and Server Hardware Do I Need for 500 Staff?
Around two GPU servers and eight to sixteen accelerators cover 500 staff, because concurrency, not headcount, sets the real hardware ceiling.
Sovereign On-Premise AI Versus Enterprise Search Assistants: Where Does Your Data Actually End Up?
With sovereign on-premise AI your data never leaves your hardware, while enterprise search assistants copy every indexed document into a vendor cloud.
How Telecoms Operators Meet the Telecommunications Security Act With AI That Never Leaves the Network
Running AI entirely inside the security-critical boundary keeps configuration and signalling data within operator control, aligning with the code of practice by design.
Only 15 Percent of Companies Can Run Agentic AI: Close the Readiness Gap Without a Consultancy Bill
Run the readiness pass on your own machines, then keep the system it stands up.