Most of what Silbad does happens inside banks, central banks, prosecutor's offices and state institutions. Those systems are closed by design — banking secrecy and personal-data protection demand it. But the principles those systems are built on, and the limits we accept on what we can do with them, belong in public.
This page is for you if you are not a customer of Silbad and never plan to be — but you live in a country whose banks, courts or markets we work with, and you would like to know how that work treats you.
Two facts about our work look like a contradiction at first, and they are not. They are the architecture.
When a bank screens a payment against money-laundering rules, neither you nor we can read the screening. That is not secrecy for its own sake — it is the thing that makes the rest of banking possible. If criminals could see what triggers a flag, none of the flags would catch anything. If your neighbour could read your screening, banking secrecy would be over.
The closure is also why our products are sold only to specific institutional buyers — banks, central banks, prosecutors, EU institutions — and never to the general public, never to advertisers, never to data brokers. There is no public version, and there is not going to be one.
The architecture, the limits we accept, the regulators we are accountable to and the ethical line we will not cross — those are not things you have to take on faith. They are written down here. They are written in our contracts with banks and states. They are written in EU law that applies to us whether we like it or not.
The rest of this page is what "open by commitment" means in concrete terms — for your data, for the decisions that affect you, for who we will and will not work with, and for what good it is supposed to do for ordinary people.
The shortest honest answer to "what does Silbad do with my personal data?" is: we do not have it. The longer answer is below.
Silbad's banking products run inside the bank — on the bank's servers or in the bank's private cloud. Your transactions, your name, your account history are processed there, the way they always were. They do not travel to Silbad.
What does travel between the bank and Silbad's central infrastructure is purely mathematical: model improvements, statistical signals, anonymous patterns. None of those reveal who you are or what you bought. The boundary is not a promise — it is the actual architecture.
Silbad is a product line of Lile, an Austrian venture regulated under EU law. The General Data Protection Regulation (GDPR), banking secrecy and the protection of business confidences are our baseline, not a marketing claim. Where a country has stricter standards than the EU floor, we honour those too.
Silbad has no consumer product. We do not run ads, we do not sell data, we do not enrich profiles. We are paid by the institutions we serve — banks, central banks, EU institutions — under proper procurement. That is the entire revenue model.
We sell our institutional products to entities operating inside the EU, the EEA and the markets in economic community with them. We do not sell to companies or institutions in countries whose human-rights situation we judge unstable, where our tools could be turned against the very people they are supposed to protect.
Before any contract is signed and before any knowledge transfer happens, Silbad conducts an ethical review of the buyer and the use case. If the review is negative, the sale does not happen — regardless of revenue.
Silbad's products produce signals, priorities, evidence chains and reasoned explanations. None of them takes an operational decision about you on the institution's behalf — by deliberate design, written into the architecture and required by EU law.
When a Silbad system raises a flag on a payment or a transaction, what happens next is not automatic punishment. The flag goes to a human analyst at the bank or the authority — usually with the reasoning, the evidence and the relevant history attached. That person makes the call. They can dismiss the flag if it does not hold up, escalate it for review, or take action under the rules they already follow.
If something goes wrong, there is also a human who is accountable. There is no "the algorithm did it" defence — for them or for us.
Under the EU AI Act (2024), systems used for fraud detection and similar financial-defence purposes are classified as high-risk AI systems. The law requires continuous human oversight on every consequential decision (Article 14), explainability (Article 13) and traceability of every decision back to its inputs.
Article 22 of the GDPR adds, separately, that you have the right not to be subject to a decision based solely on automated processing. Silbad's architecture is built so that this is not just respected — it is the default. There is no way to wire up a Silbad product to make a decision without a human in the loop.
Money laundering, terrorism financing, systemic financial attacks and large-scale corruption are not abstract problems happening on a different planet. They make ordinary life worse, in concrete and measurable ways.
Global financial-crime compliance costs are estimated at over USD 274 billion a year. Most of that is built into ordinary banking fees, and the unrecovered fraud losses are written off against tax — meaning everyone who pays for banking or who pays tax pays a share. Better defence reduces that burden.
The "money" in money laundering does not appear from nowhere. It is the proceeds of fraud against pensioners, of human trafficking, of drug trade, of corruption that diverts public funds away from hospitals, schools and infrastructure. Detecting that money is detecting the harm at its financial trace.
When a national currency, a banking system or a strategic market is attacked deliberately, the consequences — sudden price moves, frozen credit, lost savings, unemployment — fall on people who never participated in the financial event at all. SFMA defence is not exotic geopolitics; it is part of looking after the economic ground people stand on.
The deepest defence against being exploited financially is understanding how finance works. Voluntary contributions made through our Xanthos & Balius testing tools are dedicated to financial-literacy education for children — so that the people most easily targeted by fraud and predatory finance are also the people most prepared for it.
These are the questions that come up most often in conversations with people who are not banking professionals — and the honest answers, as short as they fit.
Silbad's systems flag transactions and patterns, not people. If your bank's Silbad-equipped system raises a flag in connection with a payment of yours, that flag goes to a human at the bank, who reviews it. False positives happen in any system; the human review and the explanation alongside each flag are what keep them from turning into wrong decisions about you.
If you ever have reason to believe an automated decision was taken about you without human involvement, GDPR Article 22 gives you the right to object and to have the decision reviewed. Your bank, not Silbad, is the controller of your personal data and the right contact in such cases.
No. Your bank's data does not travel to Silbad in the first place — it stays inside the bank. Silbad is a product line of Lile, an Austrian venture regulated under EU law, and we sell into the EU/EEA market. We do not maintain commercial relationships with parties outside that frame for our institutional products, and we certainly do not act as a conduit for them.
No. Surveillance, in the relevant sense, would mean Silbad collecting data about populations and analysing it centrally. The federated architecture is specifically designed so that this is impossible: the data stays where it already was, and what crosses the boundary is mathematics, not records.
What our products do support is the existing, lawful work of banks and authorities to detect and investigate financial crime — work they were already doing, under regulation that pre-dates us. We make it more effective and better evidenced; we do not extend its remit.
We cannot hand over data we never received. Our institutional products are designed so that the customer data stays with the customer institution. A lawful order to disclose data would be served on the bank, central bank or authority that holds it — not on us, because we do not hold it. We comply with lawful orders served on Silbad itself, of course; we just do not have the data those orders would normally seek.
Silbad is the financial-supervision product line of Lile, an Austrian research-driven engineering venture. It is built by an interdisciplinary team of researchers — physicists, mathematicians, philosophers, psychologists, computer scientists, economists and sociologists — who came together specifically to build serious financial-crime defence. Silbad is not a front for a bank, a state agency or a foreign technology vendor: it belongs to Lile and to no one else.
Yes. If you have a question or concern about how a Silbad system is treating you, the most useful starting point is your bank or the institution that interacts with you directly — they are the ones who can look at your specific case. If your concern is with Silbad itself or with the broader principles on this page, please use the contact channel below and we will route your message to the right people.
Public engagement is part of our work, not an afterthought. If you read this page and a question is not answered, an answer is unclear, or a commitment looks thinner than it should — please tell us. We read every message and we update this page when we are wrong.
Send us a question →For questions about a specific transaction or a specific case at your bank, please contact the bank directly — they are the institution accountable for your personal data and the only party with access to your records.
