What is adverse media screening?
Adverse media screening is the check that scans open sources, news, court records, regulatory notices and other public reporting, for negative information about a customer or counterparty that a watchlist may not yet carry. It catches risk before it is formally designated: an arrest, an indictment, a regulatory penalty, a credible allegation. The problem is that the open web is enormous and mostly irrelevant, so the discipline is as much about filtering noise as finding signal. This guide explains what adverse media screening is, why it is so noisy, how relevance filtering works, and where it sits alongside sanctions and PEP screening.
Adverse media screening, also called negative news screening, is the process of checking a customer or counterparty against public sources for negative information, from criminal allegations and regulatory action to credible reporting of financial crime. It exists because risk does not wait to be listed. A party can be under investigation, indicted or credibly accused long before any sanctions or PEP list carries them, and sometimes the conduct never produces a listing at all. Adverse media is how a program catches that exposure, which makes it a required part of due diligence rather than an optional extra, but it is also the noisiest screening check by a wide margin.
What is adverse media screening?
Adverse media screening looks beyond structured watchlists to the open record. Its sources include mainstream and local news, court and litigation records, regulatory and enforcement notices, sanctions-adjacent reporting, and in some programs leaks and investigative journalism. The purpose is to surface negative information relevant to money laundering, terrorist financing, fraud, corruption, sanctions evasion and related predicate offences. Unlike sanctions screening, which compares a party to a defined list of designations, adverse media has no fixed universe to match against; the "list" is effectively the searchable public web. That is its strength, because it reaches risk that no authority has yet formalised, and also its central difficulty, because an unbounded source set means the raw results are dominated by material that is either about a different person or not adverse at all.
Why is adverse media so noisy?
Three things make adverse media the hardest screening check to run cleanly. First, scale: the open web is effectively unlimited, and a common name can return thousands of articles that have nothing to do with the customer. Second, ambiguity: a news story rarely carries a date of birth or a passport number, so tying an article to a specific individual is a genuine identity problem, not a lookup. Third, relevance: much negative-sounding coverage is not adverse in a compliance sense, a business dispute, a critical review, an unrelated namesake accused of something, and a program that treats every negative hit as a risk event buries its reviewers. The consequence is that adverse media generates a high volume of alerts where the vast majority of alerts are false positives, and the value of the check lives almost entirely in how well that noise is filtered.
How does relevance filtering work?
Relevance filtering is the discipline that turns raw open-web hits into reviewable risk. It works on two questions at once. The first is identity: is this article actually about the customer, or about someone who shares the name? That is entity resolution, applied to unstructured text, weighing the name against any corroborating detail the source and the customer record provide, such as employer, location, age or role. The second is category: is the reported conduct relevant to financial crime risk, and how serious is it? Classification sorts hits into categories, such as fraud, corruption, sanctions evasion, violent crime or regulatory action, and grades them, so a reviewer sees an indictment for money laundering separated from a decade-old parking dispute. Only material that passes both filters, the right person and the right kind of risk, should reach a human. Everything else is noise that the program can defensibly set aside.
Is adverse media screening mandatory?
Adverse media is not always named as a standalone legal requirement, but it is treated as an expected component of customer due diligence and enhanced due diligence under the risk-based approach that global standards set. FATF guidance frames due diligence as an ongoing obligation to understand a customer's risk, and negative news is one of the clearest signals that a customer's risk profile has changed. Supervisors expect firms to consider adverse media particularly for higher-risk customers and for politically exposed persons, where enhanced due diligence applies. In practice this means a program cannot rely on sanctions and PEP lists alone and claim it has understood its customers; where credible negative reporting exists, an examiner will expect the firm to have found it, assessed it, and recorded the decision. Whether it is written as a rule or an expectation, skipping it is a defensible-diligence gap.
How does it fit with sanctions and PEP screening?
Adverse media, sanctions and PEP screening answer different questions and should be read together. Sanctions screening asks whether a party is on a prohibition list and, if so, bars or freezes dealings. PEP screening asks whether a party holds a public position that raises corruption risk and, if so, triggers enhanced due diligence. Adverse media asks what the public record says about a party's conduct, which may confirm a listing, foreshadow one, or reveal risk that will never be formally listed. The three overlap usefully: adverse media on a PEP sharpens the enhanced due diligence; a negative-news hit can be the first sign of activity that later becomes a sanctions designation. The difference in how sanctions and PEP checks are handled is set out in sanctions screening vs. PEP screening, and all three sit within the complete guide to AML screening.
How does BriteBase handle adverse media?
Our screening engine treats adverse media as a filtering problem first. Rather than returning raw open-web hits for a reviewer to sift, agentic entity resolution ties reporting to the right party using the corroborating detail on the customer record and in the source, and category classification grades each hit by the type and seriousness of the risk. That is designed to reduce false positives by up to 80% so reviewers spend their time on material that is both about the customer and genuinely adverse. Every hit carries its source citation, so the decision to escalate or set aside is evidenced in the case file for an examiner. Adverse media runs alongside sanctions and PEP screening on the same engine, and the underlying data is part of the sanctions, PEP and adverse-media data layer, available behind an existing stack or through the BriteBase adverse media screening workflow.
FAQ
What is adverse media screening?
Adverse media screening, also called negative news screening, is the process of checking a customer or counterparty against public sources for negative information relevant to financial crime. Its sources include news, court and litigation records, regulatory notices and investigative reporting, and the risks it looks for range from fraud and corruption to money laundering and sanctions evasion. It matters because risk does not wait to be listed: a party can be under investigation, indicted or credibly accused well before any sanctions or PEP list carries them, and some conduct never produces a listing at all. The consequence is that a program relying only on structured watchlists has a blind spot, and adverse media fills it. The trade-off is noise, because the open web is vast and full of namesakes, so the value of the check depends heavily on how precisely irrelevant material is filtered out before a human reviews it.
Why does adverse media produce so many false positives?
Three factors make adverse media the noisiest screening check. Scale is the first: the open web is effectively unlimited, and a common name can return thousands of unrelated articles. Ambiguity is the second: news stories rarely include a date of birth or an identifier, so linking an article to a specific customer is an identity judgement rather than a lookup, and namesakes are easily confused. Relevance is the third: much negative-sounding coverage is not adverse in a compliance sense, a business dispute or a critical review is not a financial-crime risk. Put together, these mean the vast majority of alerts are false positives, and a program that treats every negative hit as a risk event overwhelms its reviewers. The consequence is that adverse media only delivers value when relevance filtering, both identity resolution and category classification, does the heavy lifting before anything reaches a human for a decision.
How does relevance filtering reduce adverse media noise?
Relevance filtering answers two questions before a hit reaches a reviewer. The first is identity: is this article about the customer, or about a different person who shares the name? That is entity resolution applied to unstructured text, weighing the name against corroborating detail such as employer, location, age or role that appears in both the source and the customer record. The second is category: is the reported conduct relevant to financial crime, and how serious is it? Classification sorts hits into categories such as fraud, corruption or regulatory action and grades them by severity. Only material that passes both filters, the right person and the right kind of risk, should reach a human. The consequence is that reviewers spend their time on genuine exposure rather than sifting namesakes and irrelevant disputes, and the material they set aside is set aside on a recorded, defensible basis rather than at random.
Is adverse media screening a legal requirement?
Adverse media is not always written as a standalone rule, but it is treated as an expected part of customer due diligence and enhanced due diligence under the risk-based approach that global standards set. FATF frames due diligence as an ongoing obligation to understand a customer's risk, and negative news is one of the clearest signals that a risk profile has changed. Supervisors expect firms to consider adverse media especially for higher-risk customers and for politically exposed persons, where enhanced due diligence applies. In practice a program cannot rely on sanctions and PEP lists alone and claim it has understood its customers; where credible negative reporting exists, an examiner will expect it to have been found, assessed and recorded. The consequence of skipping it is a defensible-diligence gap that surfaces in examination, so most mature programs treat adverse media as required whether or not their local rule names it explicitly.
How is adverse media different from sanctions and PEP screening?
The three checks answer different questions. Sanctions screening asks whether a party is on a prohibition list and, if so, bars or freezes dealings with them. PEP screening asks whether a party holds a public position that raises corruption risk and, if so, triggers enhanced due diligence. Adverse media asks what the public record says about a party's conduct, which may confirm a listing, foreshadow one, or reveal risk that is never formally listed. The key structural difference is that sanctions and PEP screening match against defined lists, while adverse media has no fixed universe, its source is the searchable open web. That is why it reaches risk the lists miss and why it is so much noisier. The consequence is that the three are complementary rather than interchangeable: read together, adverse media on a PEP sharpens due diligence, and a negative-news hit can precede a later sanctions designation.
Sources
Cut the noise out of negative news.
Book a demo and we will show you adverse media screening with entity resolution and category classification applied, so reviewers see the right party and the right risk, each hit cited to its source. Designed to reduce false positives by up to 80%.
Book a demo