The Dual-Buyer Thesis: Selling to Regulated Firms While Licensing the Stack

Two buyers, one substrate

Most AI companies pick a side. They either sell software to end customers, or they sell the underlying technology to other vendors. Mickai does both, deliberately, because the same thing that makes our product valuable to a regulated bank makes our patent estate valuable to a hyperscaler. That is the dual-buyer thesis, and it is the spine of how we go to market.

The first buyer is the regulated firm. The second buyer is the platform that wants to serve that firm but legally cannot reach it. We serve the first directly. We position to license the second. One body of intellectual property, two revenue motions, no conflict between them.

Buyer one: the regulated firm that cannot use public-cloud AI

There is a large class of organisations that would adopt frontier AI tomorrow if the law allowed it, and the law does not. A bank governed by PRA SS2/21 cannot let model risk and customer data drift outside its own control. A hospital bound by the NHS Data Security and Protection Toolkit cannot push special-category patient data to a public endpoint. A defence supplier under ITAR and EAR cannot export controlled technical data into a shared cloud. UK GDPR special-category rules, the EU AI Act high-risk obligations, the NIS Regulations, and the extraterritorial reach of the US CLOUD Act all point the same direction.

This is not a niche. Roughly 0.85 million UK businesses, around 15 percent, sit inside this perimeter, alongside an estimated 5 million across the EU. The sovereign AI market reflects it, growing from about USD 40 billion in 2025 toward an estimated USD 148 billion by 2032.

Mickai is the sovereign AI operating system, a SIOS. It is AI that regulated businesses own and run inside their own walls, on-premise and air-gapped, with every action written to a tamper-evident, post-quantum-signed audit record we call the OAR. It is built and live, not a concept. The work for buyer one is not to invent the capability. It is to deliver it into estates that have been told for years that they must choose between modern AI and their compliance posture. We remove that choice.

The product surface is already broad. Greek-named Studio modules cover the regulated functions that matter most: Nemesis for fraud and AML, Plutus for finance and FP&A, Tyche for underwriting, Prometheus for forecasting, Iris for customer service, Nomos for compliance, Astraea for legal, Panacea for clinical, Pythia for business intelligence, and Aletheia for audit. Around them sit Trust Agent, the AMT marketing system, Vinis voice, and OAR-as-a-Service. Each one runs inside the customer perimeter, and each one writes to the same signed record.

Buyer two: the hyperscaler that needs to reach those same firms

Here is the part that is usually missed. The hyperscalers want the regulated market too. Microsoft, AWS, NVIDIA, Google, Adobe, and IBM all have sovereign and confidential-compute initiatives, because their largest enterprise customers are asking for exactly what buyer one needs. The gap is not appetite. The gap is the patented machinery that makes owned, air-gapped, fully audited AI work end to end.

That is where Mickai's estate sits. We hold 104 filed UK patent applications with roughly 2,340 claims, owned by Mickai LTD, with Micky Irons as inventor. These are filed, not granted, and the distinction matters: filing establishes priority and builds a prior-art moat. Our internal mapping identifies 196 companies and 311 patent-company pairs as potential licensees. To be precise about what that figure is, it is a sizing of where our claims plausibly read onto the direction the platforms are heading. It is potential-licensee sizing, not an allegation that anyone is infringing today.

This is why our posture toward the hyperscalers is ally, not rival. We are not building an OpenAI killer. We are building the sovereign layer that the public-cloud platforms struggle to build natively, because their architecture starts from shared multi-tenant infrastructure and ours starts from the customer's own walls. If a platform wants to offer credible sovereign AI to a defence ministry or a tier-one bank, a licence to the relevant Mickai claims is a faster path than a multi-year internal rebuild. We win when they reach more regulated customers, not fewer.

Why the two motions reinforce each other

Selling directly to regulated firms is what proves the estate. Every deployment inside a bank or a hospital is evidence that the patented approach works under real compliance load, with real audit requirements, at real scale. That evidence is precisely what makes the licensing conversation with a hyperscaler serious rather than theoretical.

Run it the other way and it holds too. A licensing relationship with a major platform validates the technology for the conservative regulated buyer who wants to know the approach is durable before committing. The direct business feeds the licensing business, and the licensing business reinforces the direct business. Neither motion cannibalises the other, because the platforms do not sell air-gapped sovereign operating systems to end customers, and we do not sell shared public cloud.

As a dated third-party signal of momentum, Micky Irons ranked number four on Crunchbase by CB Rank Person, verified live in June 2026, with the Mickai company profile sitting in the top one to two percent globally. We read that as external interest catching up with the thesis, a point-in-time signal rather than a permanent claim. We are a UK company, with Birmingham manufacturing secured, and we are building to scale.

What this means for the next phase

The dual-buyer thesis is not a hedge. It is a recognition that the same regulatory reality creates two distinct demand curves at once, and that a single coherent body of work serves both. We sell sovereign AI to the firms the public cloud cannot lawfully reach, and we offer the patented stack as a licensing layer to the platforms that want to reach them. Ally, not rival, on both fronts.

A pre-seed window is open to a small number of selected partners as Mickai scales. This is an opportunity to get involved early in a company executing both motions at once, not a sign of need. If you operate in regulated AI, hold a stake in the hyperscaler ecosystem, or simply want to understand the estate in detail, the door is open now.

To talk, write to me at micky@mickai.co.uk.

Micky Irons, founder and CEO of Mickai.

FAQ

What is the dual-buyer thesis? It is Mickai's strategy of serving two buyers with one body of intellectual property. The first buyer is the regulated firm that buys our sovereign AI operating system directly and runs it inside its own walls. The second is the hyperscaler that wants to reach those same regulated customers and can license the underlying patented stack rather than rebuild it. The two motions reinforce each other rather than compete.

Why can regulated firms not just use public-cloud AI? Law and regulation forbid it in many cases. PRA SS2/21, UK GDPR special-category rules, the NHS Data Security and Protection Toolkit, the EU AI Act high-risk regime, ITAR and EAR, the NIS Regulations, and the US CLOUD Act all restrict where sensitive data and model risk can sit. Roughly 0.85 million UK businesses and an estimated 5 million across the EU fall inside this perimeter.

Is Mickai competing with the hyperscalers? No. Mickai positions as an ally, not a rival. The public-cloud platforms want the regulated market, but their shared multi-tenant architecture makes credible air-gapped sovereign AI hard to build natively. Mickai's 104 filed UK patent applications cover that machinery, so a licence can be a faster path for a platform than an internal rebuild.

What is the status of the patents? Mickai LTD holds 104 filed UK patent applications with roughly 2,340 claims, inventor Micky Irons. They are filed, not granted. Filing establishes priority and builds a prior-art moat. Internal mapping identifies 196 companies and 311 patent-company pairs as potential licensees, which is a sizing exercise and not an allegation of infringement.

Frequently asked questions

What is the dual-buyer thesis?

It is Mickai's strategy of serving two buyers with one body of intellectual property. The first buyer is the regulated firm that buys our sovereign AI operating system directly and runs it inside its own walls. The second is the hyperscaler that wants to reach those same regulated customers and can license the underlying patented stack rather than rebuild it. The two motions reinforce each other rather than compete.

Why can regulated firms not just use public-cloud AI?

Law and regulation forbid it in many cases. PRA SS2/21, UK GDPR special-category rules, the NHS Data Security and Protection Toolkit, the EU AI Act high-risk regime, ITAR and EAR, the NIS Regulations, and the US CLOUD Act all restrict where sensitive data and model risk can sit. Roughly 0.85 million UK businesses and an estimated 5 million across the EU fall inside this perimeter.

Is Mickai competing with the hyperscalers?

No. Mickai positions as an ally, not a rival. The public-cloud platforms want the regulated market, but their shared multi-tenant architecture makes credible air-gapped sovereign AI hard to build natively. Mickai's 104 filed UK patent applications cover that machinery, so a licence can be a faster path for a platform than an internal rebuild.

What is the status of the patents?

Mickai LTD holds 104 filed UK patent applications with roughly 2,340 claims, inventor Micky Irons. They are filed, not granted. Filing establishes priority and builds a prior-art moat. Internal mapping identifies 196 companies and 311 patent-company pairs as potential licensees, which is a sizing exercise and not an allegation of infringement.