Vivox AI

The homepage documents read side enrichment only: agents access and analyze corporate registry data across 150 plus jurisdictions, scan 100 plus global sources in real time, and integrate with leading data providers to resolve screening alerts. The TransferMate rollout post names integrated registry and screening infrastructure as a delivered component.

No data provider, registry, case management system or core banking platform is named anywhere on the site. No page documents the agent writing a decision back into a customer system or taking an authenticated action outside its own dashboard. In short, enrichment across unnamed providers is documented; a named catalog and outbound actions are not.

SourceVivox AI, vivox.ai homepage and transfermate rollout postread 2026-09-07

The TransferMate case study documents agents breaking business verification into atomic, traceable steps that mirror how an analyst works, with complete CDD reports generated end to end.

The homepage describes automated KYB and KYC reviews running from onboarding workflows through risk assessment across multiple data sources, with eight named agent functions composed into a case. That is multi step, multi agent execution of a whole compliance case.

Even so, no named runtime, no BPMN and no customer facing workflow authoring surface is documented, so the orchestration is the vendor's fixed pipeline rather than one the buyer can edit.

SourceVivox AI, vivox.ai TransferMate case study and homepageread 2026-09-07

Every grounding claim on the site is external sourcing assembled per case: corporate registries across 150 plus jurisdictions, sanctions and PEP lists through leading data providers, multi language real time scanning of 100 plus global sources for adverse media, and analysis of incorporation and UBO documents supplied for the case.

What is missing is persistence. No page documents a maintained index, graph or embeddings layer over the customer's own knowledge, policies or procedures that persists between cases and stays queryable. Context is assembled fresh for each case from many external sources.

SourceVivox AI, vivox.ai homepageread 2026-09-07

The vendor's TransferMate rollout post names human in the loop controls for operational oversight as one of three delivered components, alongside the agents and the registry infrastructure. The case study says every step the agents take is explainable and fully documented, so the compliance team kept full oversight.

AI approval rates on agent outcomes rose from 60 to 85 percent, which means the vendor ships and measures its own per output approval step. One gap: the approval screen appears in dashboard screenshots but is not described in text, so where the analyst approves is inferred from the metric rather than read from a feature page. Supervised feedback from analysts is a learning loop, not a gate.

SourceVivox AI, vivox.ai TransferMate rollout post and case studyread 2026-09-07

Not known. Neither certifications nor access controls can be read on any reachable first party page. The homepage states governance aligned with FCA requirements, the EU AI Act and global standards, with full audit trails; that describes governance and logging, not an attestation or an access control surface. The wall of regulator logos shows the customers' regulators, not certifications held.

No SOC 2, ISO certificate or trust center is named on the homepage, resources index, case study or rollout post, and no SSO, SAML, SCIM, roles or permissions appear anywhere. A search for a trust or security subdomain returned only similarly named companies. The terms and privacy policy sit on a Termly policy viewer.

SourceVivox AI, vivox.ai homepage, resources, case study, rollout post and subdomain proberead 2026-09-07

The TransferMate case study documents that agents break verification into atomic, traceable steps, with every decision logged and auditable and every step explainable and fully documented. The homepage names explainable AI with full audit trails and explainable decisions as the product's design center.

A per step record of what the agent did, plus a per decision explanation of why, lets a team reconstruct why an outcome happened, not just see that it did. The logs are the vendor's own and do not depend on the customer's systems. A hosted customer testimonial adds that outputs were validated during clients' external audits.

SourceVivox AI, vivox.ai TransferMate case study and homepageread 2026-09-07

No page documents a memory layer with a stated scope and a stated lifetime. A self learning agent that improves through analyst feedback is learning absorbed into the model, not memory a buyer can inspect. Case and periodic review state is the application's own data model: deleting it would delete the compliance record, not a memory. No chat interface is documented, so there is no session memory to describe either. The site does not say where memory lives or how to delete it for one customer.

SourceVivox AI, vivox.ai homepage and case studyread 2026-09-07

Not known. No reachable first party page names a hosting region, a customer environment option or a way to choose one. The homepage and rollout post describe customers operating across Europe, the US and APAC, which is where the customers are, not where the data is held. Deployment to customers in more than 100 countries is a sales footprint, not a hosting choice. The privacy policy and terms sit on a Termly policy viewer. Reading the privacy policy in a browser, or asking Vivox AI directly, would settle where customer data is held.

SourceVivox AI, vivox.ai homepage, rollout post and Termly privacy policy attemptread 2026-09-07

The homepage describes pre trained AI agents under the Rachel workforce name and lists eight named functions: Adverse Media Monitoring, Sanctions and PEP Alert Reviews, AI Website Analysis, Corporate Registry Data Analysis, Ongoing Monitoring, Incorporation and UBO Docs Analysis, Narrative Report Writing, and Request for Information. Each unit stands on its own: without adverse media monitoring, sanctions alert review still does its whole Level 1 job.

The vendor's case studies show units adopted alone: Osome on adverse media screening, TELF on adverse media due diligence, TransferMate on KYB and EDD. That is a choice among prebuilt agents rather than tiers of one product. The counter reading: the homepage labels the list as customized skills of one AI compliance analyst persona, there is no catalog page, and whether each unit is licensed separately is not stated.

SourceVivox AI, vivox.ai homepage and case studies indexread 2026-09-07

Work reaches the agents three ways on the homepage: an onboarding case arriving for KYB or KYC review, a screening alert to be resolved, and an Ongoing Monitoring function running 24/7 with real time scanning of 100 plus sources, plus periodic reviews. That is event driven intake, described as a claim. No page documents a configurable trigger surface, a schedule the customer sets, a webhook or an inbound queue, so how a case or alert enters the system is not stated.

SourceVivox AI, vivox.ai homepageread 2026-09-07

The only first party statement about models is a hosted testimonial on the homepage describing precision driven atomic AI models fine tuned by expert humans in the loop. Those are the vendor's own fine tuned models: no model provider is named, and no customer or admin choice of model is documented anywhere. Only the vendor chooses. Fine tuning its own models is neither routing across providers nor buyer choice. A third party write up describes a multi model ensemble, but that is not on the vendor's own pages.

SourceVivox AI, vivox.ai homepageread 2026-09-07

Nothing documented. No developer portal, API reference, SDK, MCP server or agent card is linked from any page on the site. The footer carries only case studies, news, resources and the Termly legal links, and a search for developer documentation on vivox.ai returned nothing. Agents connecting to data sources is the platform reaching out to other systems, not an interface developers can build on. A name collision to watch: the Vivox Core SDK on docs.unity.com is Unity's gaming voice and text chat product, not this company.

SourceVivox AI, vivox.ai homepage and footer plus developer proberead 2026-09-07

The evaluation evidence on the vendor's pages is a live operational metric: AI approval rates rose from 60 to 85 percent during the TransferMate deployment, through a structured feedback loop in which analysts working directly with the agents improved decision quality. That is a supervised ramp on live operations.

No test harness, scored test cases, release gate or controlled post deployment experiment is documented. A hosted testimonial that outputs were validated during clients' external audits describes the customer's audit of its own compliance program, not an evaluation of the agent, and it is a customer quote.

SourceVivox AI, vivox.ai TransferMate case study and rollout postread 2026-09-07

The homepage names AI Website Analysis as one of eight agent functions: reading the subject business's own website at case time as part of a KYB review. That is a run time fetch of a web page while doing the work, not a crawler building an index, and the mechanism (headless fetch or scrape) is not stated. The multi language real time scanning of 100 plus global sources for adverse media is monitoring rather than browsing. No hosted or local browser, desktop session or remote computer control is documented.

SourceVivox AI, vivox.ai homepageread 2026-09-07