T57

We produce enough food in the world. That's not the problem. The problem is that a surplus in one country and a deficit in another don't automatically connect. A producer with grain and a government that needs grain can sit on opposite sides of a gap that has nothing to do with production.

I call it “strategy without execution” because I kept watching the same pattern across the countries and institutions I was working with. A country signs a food-security strategy, a producer has supply, a bank has capital, a logistics company has capacity, and none of it moves, because each of those sits in its own silo. An MoU doesn't move a tonne of wheat. A marketplace listing doesn't finance a shipment. A logistics platform doesn't verify a buyer's creditworthiness. Somebody still has to connect verified supply to verified demand, arrange the financing, handle the logistics, clear compliance, and settle the trade. That connective layer is what was missing.

That's the specific gap T57 is built to close. Not another strategy document, but the execution infrastructure underneath the ones that already exist.

I'd rather be precise about this than impressive about it.

Right now we have a live pilot running across five countries, and 34 shipments have actually gone through the ecosystem: real transactions, real commodities moved, not simulations. Alongside that we have relationships across 67 countries and more than 150 MoUs and LOIs tied to the broader T57 ecosystem.

Those are two different things and I don't like collapsing them into one number. An MoU is an agreement. A shipment is proof. The five-country pilot matters because it's where we're stress-testing the whole chain, sourcing, verification, transaction coordination, physical movement, in the real world, not on a slide.

The next phase is scaling what's been validated in that pilot into more markets and corridors. Commercial launch is planned for February or March 2027. So today T57 is an operationally emerging platform with a working pilot and proven shipments, not a fully live global ecosystem. I'd rather say that plainly than let the architecture description imply more than it currently is.

I'd split this into two numbers rather than one, because combining them would overstate things.

We have engagement across 67 countries: governments, institutions, producers, trade bodies, companies. Separately, we have more than 150 MoUs and LOIs across the ecosystem. Those aren't the same measure. The 67 is the breadth of the network. The 150-plus represents formalised cooperation, interest or commercial intent across the broader T57 ecosystem, and even within that group the maturity varies a lot. Some are institutional relationships, some are commercial leads, some are moving toward pilots, and a smaller number are closer to actual transaction activity.

An MoU is not a transaction. I've said that internally more times than I can count, and I'll say it here too.

What actually matters is conversion: how many of those relationships turn into verified supply, verified demand, financed and executed contracts, and eventually recurring trade. So rather than round everything up into “70+ government partnerships,” I'd rather give you the honest 67-country and 150-plus-MoU numbers and let the execution data prove what they're worth over time.

I won't claim the AI has already independently caused a measurable allocation of commodities at scale. That would be overstating where we are.

What we're building is different from a standard forecasting tool. A conventional system tells you “Country X is likely to face a wheat shortage.” What we want is for the system to eventually say something closer to this: this country has a verified requirement of this quantity and spec, these producers can plausibly meet it, the supply is verified, the route is viable, the financing works, compliance is satisfied, execute. That's the gap between food intelligence and food-trade execution, and we're still building toward the execution end of it.

On the data problem, you're right that smallholder-heavy markets are exactly where ground data gets thin, and I don't think the answer is telling the model to fill in the blanks. We layer verification instead: producer-reported data, partner and institutional datasets, trade records, remote sensing where it's useful, and human or institutional verification when the primary data isn't good enough. The rule we hold to is simple. The system should know when it's uncertain and ask for more verification rather than manufacture confidence from weak inputs, especially on something with the stakes of food security.

Yes. If we built it wrong, T57 could absolutely become a strategic vulnerability itself. Any infrastructure that becomes critical to food security creates concentration risk, and I don't want to replace today's fragmented system with a new single point of failure that happens to be called T57.

So the design principle is distributed trust, not centralized dependency. A government shouldn't have to give up control of its food data just to participate. National data and sovereign intelligence stay under the jurisdiction and permissions of whoever owns them. T57 connects systems. It doesn't own them.

Which gets to the second half of your question: why should a government in sub-Saharan Africa or South Asia trust infrastructure sitting in the Gulf? Honestly, it shouldn't trust us just because of geography. Trust gets earned through governance, technical safeguards, contractual protections, and behavior over time, not by where the servers happen to sit. The Gulf gives us a useful position between Asia, Africa and Europe, with strong ties into trade, logistics and capital. But location shouldn't determine who controls another country's data. If a government can't get a straight answer from us on where its data is, who can access it, and what we're allowed to do with it, then we haven't earned that trust yet.

T57 shouldn't own data just because it passes through our infrastructure. The rights stay with the party that has the applicable legal rights to the data, whether that's a government, a company, or a producer, under applicable law and whatever contract governs the relationship. Our job is the infrastructure and intelligence layer on top, not data ownership underneath.

For sovereign data specifically, the architecture has to support keeping it in-country. If a government needs sensitive information to stay within its borders, T57 needs to be able to run that through national or federated infrastructure rather than forcing everything into one central system. Participating in a global trade network shouldn't mean a country's raw data has to leave the country.

On algorithm auditing, I want to be honest that the governance principle and the current maturity of implementation aren't the same thing yet. We believe decisions on creditworthiness, allocation, or food-security prioritization need to be auditable and independently reviewable, and that's where we're building toward. I'm not going to claim we already have a fully deployed third-party audit regime running, because we don't, and I'd rather say that than overstate it. But the line we won't cross is letting AI become an unaccountable authority over decisions that governments and institutions should be accountable for.

I'll give you honest numbers on what I can commit to today, and I'll also tell you where I can't yet, rather than inventing precision I don't have.

On the business side, our 2028 targets are 11 to 17 active markets, roughly 40,000 active subscribers, and around $127 million in revenue run-rate. Those are targets, not results we've already hit, and I want to be clear about that distinction. I'd stand behind an auditor checking us against those three numbers in 2028.

But you asked for tonnes traced, dollars saved, days reduced, farmers onboarded, and I'm not going to hand you numbers for those today that I can't defend. We don't have them yet, and I'd rather say that than manufacture a figure to sound more convincing. What I can tell you is that those are exactly the right questions, and they're the metrics that actually prove whether T57 changed anything. Markets, subscribers and revenue tell you the business is real. Tonnes moved, waste avoided, cycle time cut and producers connected tell you whether food is actually reaching people more efficiently, and that's the harder, more important number.

So my honest answer is this: by 2028 I expect to have the business metrics validated, and I expect to be able to report real figures on trade volume, producers onboarded and cycle-time reduction from the markets where we're live. If we can't produce those numbers by then, with sourcing you can check, that's a fair sign the model isn't working the way I believe it will.

The goal is to shift leverage toward producers and importing governments, away from whoever currently benefits from the information gap between them.

A smallholder producer today is often several steps removed from the buyer. They don't know where the demand is, what spec is wanted, what the market will pay, or what financing exists. On the other side, an importing government can have a real strategic need without visibility into global supply. T57 is meant to close that gap. Producers get market visibility and access, governments get planning certainty, and intermediaries get paid for the value they actually add: financing, logistics, risk management, local knowledge, instead of just sitting on information other people don't have.

The harder question is what stops T57 from becoming that same chokepoint under a different name. The answer has to be structural. Open, documented APIs. Non-exclusive participation. Data portability. Governments and companies connecting their own systems instead of being forced to replace them. Contractual protection against lock-in. Nobody should have to abandon their existing infrastructure to use T57.

We're not trying to own the food-trade ecosystem. We're trying to make it work better. So by 2035, I don't want T57 described as the company that controls global food trade. I want it described as the infrastructure that made global food trade more transparent, more executable, and more resilient.