Nominis

The load bearing work here is inference rather than lookup, which places it above most of this lane. Clustering more than a billion wallets into entities is an attribution problem solved by learned heuristics rather than by a register, and scanning more than 100,000 new wallets daily means that attribution is continuously recomputed rather than curated by hand.

The stated fusion of on chain activity with off chain intelligence from open, deep and dark web sources, behavioural analytics and geo location is a multi modal inference problem, and the company's own framing depends on it, since detecting a behavioural break that legacy tools miss is by definition a model judgement rather than a rule. References to supporting more explainable compliance decisions imply outputs that require explanation. Two things hold it out of the top band.

The flagship differentiator, the terror financing attribution database, is a curated dataset rather than a model. And across two passes nothing published names a technique, architecture or learned component anywhere.

The institution writes the logic and the workflow is built around an investigator. A custom risk policy builder lets each firm design rules reflecting its own risk profile and jurisdiction rather than inheriting the vendor's thresholds, which places the substantive decisions with the customer, and the company presents that configurability as a central differentiator rather than an advanced option.

The investigative surface reinforces it, with tools for mapping fund trails, inspecting complex flows through mixers and bridges and supporting regulator queries, all of which describe a person examining evidence. Stated emphasis on explainable compliance decisions implies outputs a human is expected to interrogate rather than accept. What is not published is a boundary.

Real time blocking of risky flows before settlement is offered, which is machine action on money movement, and nothing states what threshold governs it, whether any category of block requires review, or what a wrongly blocked counterparty is told.

Volume and speed are quantified, accuracy is not. Published figures describe throughput and coverage rather than performance: more than 70 blockchains, over a billion clustered wallets, more than 100,000 new wallets scanned daily, screening responses under 500 milliseconds, and an 80 percent reduction in investigation and decision time. Those tell a buyer the platform is fast and broad, not that it is right.

The one competitive performance claim, that the company has exclusively flagged 10 billion dollars in illicit crypto, asserts detection that named competitors missed, which is a strong assertion carrying no methodology, no sample, no definition of exclusivity and no way to verify.

Across two passes no false positive rate, no attribution accuracy measure, no recall against a known illicit set, no validation methodology and no drift or coverage change disclosure was located, which matters for a product whose clustering is recomputed continuously as new services appear.

Genuine institutional corroboration of capability, and no customer evidence at all. The corroboration is unusual and checkable: the company publishes joint research with a national security think tank tracing how a designated network converted smuggled gold into cryptocurrency and moved the proceeds through apparently innocent wallets, which is original investigative work validated by an institution with no commercial interest in the platform.

Backing from an innovation lab jointly operated by a global card network, participation in that network's accelerator, and recognition from a national innovation authority add further external assessment of the business. Against that, across two passes no customer is named anywhere, no customer count is published, and the claim of being trusted by law enforcement agencies identifies no agency.

The published figures are vendor stated and mostly volumetric: 10 billion dollars in illicit crypto exclusively flagged, more than a billion clustered wallets, 100,000 new wallets daily, an 80 percent reduction in investigation time. The company was founded in June 2023 and employs on the order of ten to fifty people.

One genuine public interest contribution and no safety documentation. The contribution is the published research with a national security institute, which puts investigative findings into the public domain rather than reserving them for customers, and the company's stated emphasis on explainable compliance decisions points in the right direction on interpretability. Beyond that the record is empty.

Across two passes no model card, evaluation methodology, red team result, incident disclosure or acceptable use boundary was located. The acceptable use gap is the notable one here rather than a routine omission, because the same platform is sold to commercial compliance teams and to law enforcement agencies, and a proprietary attribution database identifying state sponsored actors, proxy entities and facilitators is dual use by construction. Nothing published describes what uses the company declines, how it handles requests from state actors, or what separates a commercial screening deployment from an investigative targeting one.

Nothing is published, and the architecture makes that absence weigh more than it would at a chain only analytics vendor. Across two passes no privacy policy content, data processing description, retention schedule or subprocessor list was located. The specific exposure follows from the product's central claim. Analysing the public ledger involves data that is by definition public.

This platform goes deliberately further, fusing on chain activity with intelligence gathered from open, deep and dark web sources, behavioural signals and geo location, in order to link pseudonymous addresses to real world identity and place. That is assembly of profiles about people from material they never published and cannot see, sourced in part from criminal marketplaces and breach data, and applied to decisions that block their transactions. Nothing published describes what off chain material is retained, how long, how it is verified before it affects a score, or what rights a profiled individual has.

Nothing was located. Across two passes no certification, attestation report, trust centre, security page, penetration test summary, subprocessor list or security contact appears on any surface, and no framework is named anywhere in the published material. This is complete absence rather than thin or stale disclosure.

The mitigating inference available at larger vendors does not apply, because it depends on named supervised customers having run third party assessments, and this company names no customer at all. Its association with an innovation lab operated by a global card network and participation in that network's accelerator do involve external diligence, but that is commercial and product assessment rather than an information security examination, and neither is presented as the latter. The gap matters in proportion to what the platform holds, which includes off chain intelligence and behavioural profiles linking wallet addresses to identifiable people.

An unregulated supplier with unusually specific regulatory literacy for its size. Published material engages named regimes rather than gesturing at compliance: the Australian regulator's 2026 crypto anti money laundering framework and what it changes for monitoring, New York state supervisory guidance and its virtual currency licence, the international standard setter's travel rule recommendation, and European requirements, each addressed with practical guidance rather than a mention.

For a company founded in 2023 with a small team, that depth of jurisdictional coverage is real work. It confers no standing. The company holds no authorisation and claims none, and across two passes no supervisory examination outcome, regulatory review or accreditation was located. Its position inside an innovation lab jointly operated by a global card network is commercial sponsorship rather than regulatory recognition. No position on the European artificial intelligence regulation was located, despite the platform producing risk determinations about identifiable people.

This platform makes the most consequential determination in the crypto lane and publishes nothing about how it is governed. The flagship asset is a curated database attributing wallets to terror financing, covering state sponsored actors, proxy entities, facilitators and brokers, assembled partly from open, deep and dark web sources, and applied to block transactions before settlement.

Being wrongly included carries consequences well beyond a declined deposit, since a terror financing attribution can mean funds frozen, an account closed and a report filed to authorities, and the subject is never notified and is not the vendor's customer. Two governance questions follow and neither is addressed. What evidentiary standard admits an entity to the database, who reviews additions, and how is an error corrected.

And how are jurisdictional differences handled, given that designations of who constitutes a terror financier differ between governments, so an attribution reflecting one jurisdiction's view is applied to customers operating under others. Across two passes no methodology, review process, accuracy measure, appeal route or fairness analysis was located.

No commercial instrument is published and the third party exposure is the starkest in this lane. Across two passes no terms of service, master agreement, warranty, indemnity, liability cap or service level was located, and nothing states what a customer is owed if a screening result proves wrong in either direction. The uncovered party is not the customer.

Someone wrongly attributed in a terror financing database, or wrongly clustered into an entity through an attribution error, has no relationship with this company, receives no notice, and has no published route to correction, while the consequence they experience is a blocked transaction, a frozen balance or a report to authorities.

The company's own model makes this concrete, since it markets blocking risky flows before settlement, which means the affected party learns only that a transfer failed. No appeal process, correction procedure or contact route for a mis attributed subject was located.

The categories of input are disclosed more candidly than most competitors manage, and no supplier is identified. On the candour side, the company states plainly that its determinations draw on off chain intelligence from open, deep and dark web sources alongside behavioural analytics and geo location, and admitting that dark web material feeds a compliance score is disclosure many vendors would prefer to leave implicit.

The terror financing attribution database is presented as proprietary and curated in house, so a buyer knows the differentiating asset is owned rather than licensed. What is absent is specificity of a kind that matters here more than usual. Deep and dark web is a description of where material comes from, not who supplies it or how it is verified, and the reliability of intelligence sourced from criminal marketplaces and breach dumps depends entirely on that. No data supplier, intelligence partner, technique or model provider is named anywhere.

Integration is a stated differentiator with nothing published to inspect. The company positions itself explicitly against competitors whose integration timelines run in quarters rather than days, offers both a secure interface and a visual dashboard so a firm can adopt either as a screen or as an embedded check, and describes embedding transaction checks into smart contract flows and running them before settlement rather than after.

The stated response time under 500 milliseconds is a genuine integration relevant specification, since a check that fast can sit inline in a deposit or withdrawal path without the user waiting. Chain coverage above 70 networks with dedicated material per asset means a buyer can check whether their own mix is supported. What is missing is anything an engineer can evaluate before contact.

Across two passes no public interface documentation, developer portal, sandbox or code repository was located, and no exchange, custody platform, wallet provider or compliance system is named as a supported integration.

A hosted service with the placement questions unaddressed, in a company whose data footprint makes them pointed. Across two passes no hosting provider, region list, tenancy description or residency commitment was located, and no customer hosted option is described. Two factors sharpen the question.

The platform accumulates off chain intelligence, behavioural profiles and geo location signals about identifiable people alongside chain data, so what rests with the vendor is materially more sensitive than a ledger index. And the company is Israeli incorporated and headquartered while selling to virtual asset service providers under Australian, New York state and European supervision, several of which impose expectations on where regulated data is processed and which authorities may compel access to it. A regulated buyer would need those answers before onboarding and none is discoverable publicly.

Pricing accessibility is a published argument and the price is not published, which is a tension worth naming. The company's competitive case rests partly on commercial terms: it criticises established providers for quoting enterprise contracts to virtual asset service providers before those firms have enterprise revenue, and for integration timelines that run in quarters rather than days, positioning itself as the accessible alternative for fast growing crypto businesses.

Having made cost and accessibility the differentiator, it publishes no rate, unit, tier or contract term anywhere across two passes, so a buyer cannot verify the claim that this will be cheaper or check the shape of the commitment before entering a sales process. The estate spans wallet screening, transaction monitoring, forensic investigation and a policy builder, which could plausibly be charged by screened addresses, monitored transactions, seats or subscription, and nothing narrows it. The published entry route is a demonstration request.

Buyer types named by category across a genuinely wide span, with nobody named inside any of them. The stated clientele reaches beyond crypto native firms to banks, payment processors and law enforcement agencies, alongside exchanges, payment gateways, custodial services and decentralised finance protocols, and the company addresses both operational compliance and investigative casework as distinct jobs.

Chain coverage is the strongest concrete dimension, at more than 70 blockchains with dedicated material for individual assets rather than a single generic page. Regulatory coverage is similarly specific, engaging the Australian regulator's 2026 crypto rules, New York state guidance and its licensing regime, and the international travel rule standard, which indicates real jurisdictional breadth.

What is missing is evidence: no institution is named in any segment, no customer count is published, and the company is small and recently founded, so the breadth described is addressable market rather than demonstrated footprint.