Domu AI

The conversation is the deliverable. Generative agents hold the servicing call, text and email themselves, and a behavioral model decides which account to approach, at what hour and through which channel before the agent opens its mouth. Take the models away and there is no workflow product left, no dialer and no case management system underneath, only an integration into a lender's core system with nothing to send through it. The product is assistant native, not a communications platform with models bolted on, and that sets it apart from channel platforms in this category that added intelligence to an existing spine.

The oversight architecture is split across three named modules sitting at three distinct points in the execution path, and each one names the law it is checking against. Before deployment, Alex stress tests conversation flows against Fair Debt Collection Practices Act and Telephone Consumer Protection Act boundaries in a synthetic environment and confines the agent to an approved repository of data.

During the conversation, required disclosures are delivered automatically, consent state is tracked in real time, and inappropriate language, threats and policy deviations are flagged as they occur rather than found afterwards. After deployment, Jordan audits live conversations against unfair, deceptive or abusive acts and practices standards and state specific collection law, flags policy drift and assembles evidence an examiner can read.

A human supervisor receives any conversation that moves beyond preset parameters, and sign off is required on high stakes decisions. Naming the statute at each stage and placing a staffed human layer in the path is what separates this from a general assurance that guardrails exist. The preset parameters themselves are never given values: no calling window, no contact frequency cap and no confidence threshold is published.

A real validation mechanism runs at both ends of deployment. Conversation flows are adversarially tested in a synthetic environment before the agent reaches a customer, and live behavior is audited continuously afterwards. The stated output is audit ready evidence an institution can put in front of an examiner, which hands the buyer something to inspect rather than a claim to accept. None of it is quantified.

No accuracy, containment or escalation rate is published, and no false positive rate is given for the compliance screening the institution would be relying on. No retraining cadence or drift measurement appears, no ISO 42001 certification is held, and no model documentation or validation pack is offered.

Named institutions appear next to specific numbers. SBS Insurance is credited with a three percent liquidation improvement inside the first two months of use, Skandia with $1 million recovered, and BNP Paribas with a forty percent increase in right party contact rate. The company also reports a forty six percent reduction in cost per account resolved and a thirty percent fall in complaints per hundred calls at an unnamed top five fintech.

Nu, DigniFi and Alorica are named as customers separately. A $25 million Series A led by Standard Capital puts an independent party with money at stake behind the diligence. Every figure is company reported, with no customer published announcement and no third party verification, so the references are named but the measurement is not independent.

A safety architecture is named, and one part of it is a real boundary rather than a description of good intent. The agent is confined to an approved repository of data, so what it can draw on in a live conversation is bounded in advance. It is adversarially tested against threats and inappropriate language in a synthetic environment before reaching anyone, and every supported conversation is then monitored and evaluated. The boundary between clients is left open.

The behavioral models learn from institutions' historical servicing data across a customer base the company describes as including eight of the twenty largest banks and insurers in the Americas, which means competing creditors. Nothing states whether one creditor's conversations inform the models serving another, or whether a consumer's behavioral profile follows them when a different creditor places their account. Nor is it stated what happens to a client's contributed data when the relationship ends.

No privacy policy detail, processing agreement, subprocessor list or retention schedule is published. What is at stake is sensitive: recorded conversations in which consumers discuss why they cannot pay, disclose hardship circumstances and hand over payment credentials. These come from people who are behind on an obligation and who did not choose this vendor.

The company also reaches across the Americas, with named customers in Brazil and Colombia as well as the United States, so the same recordings cross data protection regimes that each impose their own obligations. Nothing addresses which regime governs a given conversation or how long the audio and transcripts are kept.

Two audited credentials are claimed, SOC 2 Type II and PCI, and both fit a platform that negotiates and takes payment on live consumer calls. The presentation is the problem. The same row lists the Consumer Financial Protection Bureau and the Telephone Consumer Protection Act as though they were certifications.

A federal supervisor and a federal statute are not schemes anyone can certify against, so a buyer reading the row cannot tell which items a third party audited and which are assertions of compliance with law. The two audited certifications still stand. No scope statement, audit period or named auditor is given, and no assessed payment card level is stated. There is no trust center or evidence portal.

The company presents itself as a platform sold to regulated institutions and holds no license in its own name in any published source, which is the correct posture for a software vendor. One outside claim is set aside. A third party application directory describes the company as a licensed collections agency as well as a platform. If the company's own material said so, that would place it inside the regulatory perimeter alongside the licensed digital first agencies in this category. The claim appears in no company published source and names no state, jurisdiction or license number, so it is not reflected here.

The published governance work is conduct screening rather than fairness testing, and the difference matters here. Checking a conversation against unfair, deceptive or abusive acts and practices standards catches what an agent said to one person; it does not test whether the system treats comparable people differently.

The exposure sits one layer earlier, in the behavioural model that decides who is contacted, how often, through which channel and with what tone, which is a differential treatment decision made about people in financial distress and made before any conversation a compliance module could review.

No fairness evaluation, differential outcome analysis or cohort level reporting appears anywhere, and nothing states whether contact intensity or channel assignment distributes evenly across geography, product or borrower type.

The architecture is preventive from end to end, with no route afterwards, which is the structural position of software vendors in this category. The creditor remains the regulated party and absorbs the statutory consequence of a call it did not script, made by a model it did not build and cannot validate, while the platform operates the agent. Evidence assembled for examiners is a defense for the institution, not a remedy for the consumer.

No liability position, error rate or remediation commitment is published. A consumer wrongly contacted, misidentified as the right party or held to a payment commitment they did not understand has no described way to have the interaction reviewed by anyone.

No model provider, family, version or hosting arrangement is named. The material describes generative agents and language models trained to sound like high performing collectors and stops there. The voice specific version of this gap is the sharper one: speech synthesis and real time transcription are components a company of this size almost always licenses rather than builds, so an undisclosed supply chain on a voice product means a consumer's spoken account of their financial difficulty passes through at least one third party the buying institution was never told about, and the institution cannot enumerate its own fourth party exposure when its examiner asks.

This is not a plug and play voice bot, and the company says so: it will not work without connection to the institution's core systems. Buyers are told to expect API setup, data mapping and agent training before the platform can handle live conversations. The behavioral layer depends on the lender's own historical servicing data, so the integration load is a stated prerequisite rather than an optional extra. The detail is missing.

No core banking, loan management or servicing platform is named, no integration count or partner directory is published, and there is no public API documentation. The depth is asserted through the size of the implementation rather than shown.

The platform is cloud delivered and integration led, and nothing beyond that is published about where it runs. No hosting region, residency commitment, tenancy model or single tenant option appears. Residency is a concrete question rather than a formality here, because a San Francisco company is handling recorded consumer collection conversations for institutions in Brazil, Colombia and the United States, and each of those markets treats the movement of that data differently. Nothing states where audio, transcripts or the behavioural profiles derived from them are stored or processed, or whether a Latin American client's borrower conversations remain in region.

No price, tier, billing basis or starting figure is published for any part of the platform. The gap is conspicuous because the company publishes extensively on how institutions should evaluate the total cost of ownership of platforms in this category, including the costs of integration work and compliance remediation, while disclosing none of its own.

A buyer cannot tell whether the model is subscription, per account, per connected minute or a share of amounts recovered, and in collections those imply materially different incentives, since a vendor paid on recovery has an interest in contact intensity that a vendor paid a licence fee does not. The stated requirement for substantial integration work also implies an implementation cost that is never sized.

Six distinct buyer types are addressed, and the named references land in several of them. The platform is sold to banks, fintech lenders and buy now pay later providers, and to insurers, loan servicers and business process outsourcers. Nu, DigniFi and Alorica are named as customers, and SBS Insurance, Skandia and BNP Paribas appear alongside published outcomes.

That set spans a digital bank, a specialist finance lender and an outsourcer, plus an insurer and a European banking group, so the breadth is real rather than one buyer described several ways. The company states it works with eight of the twenty largest banks and insurance companies in the Americas. Coverage stops at the Americas, and the material offers nothing for Europe or Asia. That limits the claim; it is not a gap in the evidence.