Adverse Media Screening demo
Agents search each language tied to the customer, group copies into one story per event and challenge each match before two people sign 1:33
What the demo shows
KYC analysts find the negative news that matters about a customer, in any language, cited
Agents search every name variant in every language tied to the customer, translate sentence by sentence and keep the original, group syndicated copies into one story per event, compare six identifiers with the customer record, and then try to prove each match is a different person before proposing anything. What survives is classified by predicate offense and stage. A KYC analyst confirms relevance with a reason code; an EDD reviewer accepts it into the customer file; the compliance manager approves any rating change.
Transcript
- Adverse Media Screening finds the negative news that matters about a customer, in any language.
- In one survey, sixty-one percent of investigators said search engines return too much irrelevant news.
- Rafael Monteiro Duarte directs a Corvane client. The search found thirty-seven articles.
- Agents group the copies: thirty-seven articles become four stories.
- A second agent proves two are other men, and a third is harmless company news.
- A Portuguese report on a procurement fraud probe names Rafael M. Duarte, fifty-two, engineer.
- He was fifty-two that day, and an engineer.
- The analyst confirms: corruption, under investigation, with a reason code.
- A reviewer accepts it into the file; a manager approves the higher rating.
- Every decision is kept five years, with an export for examiners.
- Thirty-seven articles became one cited story in minutes, signed by two people.
- Built on Botminds: agents, data, screens and governance on one platform.
The narration as written for the film. Captions are on the player.