Adverse Media Screening That's Actually Grounded

Sanctions and PEP lists tell you what regulators have already formalized. Adverse media tells you what's coming. A customer named in a fraud investigation, a corruption exposé, or a money-laundering case will often appear in the press long before — or instead of — any official list. Adverse media screening is the early-warning layer of AML: the negative-news signal that catches risk while it's still a headline.

The catch is that adverse media is also the noisiest signal in the entire stack. Names are common, news is plentiful, and a naive keyword sweep produces a flood of false positives. Didit treats adverse media as one category within its 1,300+ watchlists, scored with the same two-score model that governs the rest of AML screening — so a negative-news hit only reaches your analysts when the engine is confident it's actually your customer. It runs as a workflow step or a standalone API at $0.20 per check.

Key takeaways

• Adverse media is a category, screened alongside sanctions, PEPs, criminal records, and warnings in one $0.20 call — not a separate product.
• The two-score model applies. A Match Score decides whether a negative-news hit is really your customer; the Risk Score decides how much it matters.
• Namesake suppression. Because adverse media is so noisy, the Match Score threshold is what keeps a common-name news story out of your review queue.
• Auditable review states — False Positive, Unreviewed, Confirmed Match, Inconclusive — give you a defensible record for each hit.
• Continuously monitored. With ongoing monitoring, adverse media that emerges after onboarding is flagged the next day.
• $0.20 per check, as a workflow step or a standalone POST /v3/aml/ call.

What adverse media screening does

Adverse media screening checks your customer against negative news tied to financial crime and related conduct — fraud, corruption, money laundering, organized crime, regulatory breaches. When the engine finds a media-derived record that matches your subject, it surfaces it as a profile in the Adverse Media category, with a Match Score for identity confidence and a contribution to the overall Risk Score.

The point is not to flag everyone who has ever been in the news. It's to surface the adverse coverage — the kind that, if you onboarded the customer without seeing it, would later look like a failure of due diligence. And because the same person can appear in unrelated stories, the two-score model is what separates "this is genuinely your customer, in a corruption case" from "someone with the same name was quoted in a sports article."

Why it matters

Most modern AML regimes expect adverse-media checks as part of customer due diligence, and explicitly as part of enhanced due diligence for higher-risk customers and PEPs. The reasoning is that official lists lag reality — a person under investigation may not be sanctioned or convicted for years, but the risk is present the moment the investigation is public. Skipping adverse media means accepting customers whose risk is visible to any journalist but invisible to your screening.

The reason adverse media is so often done badly is the false-positive problem, amplified. Sanctions lists are curated and structured; news is unstructured, vast, and full of common names. A screening approach that can't reliably tie a story to your customer either floods analysts with irrelevant articles or gets switched off. The whole value of grounding adverse media in the two-score model is that it inherits the same identity-confidence discipline as sanctions screening — so the signal is usable instead of overwhelming.

Technical details

Adverse media is screened by the standard AML check; no separate endpoint or extra call.

curl -X POST https://verification.didit.me/v3/aml/ \
  -H "x-api-key: $DIDIT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "first_name": "Daniel",
    "last_name": "Okafor",
    "date_of_birth": "1979-06-22",
    "country": "NG"
  }'

An adverse-media hit comes back as a categorized profile with its own Match Score and review state:

{
  "aml_status": "In Review",
  "risk_score": 66,
  "matches": [
    {
      "profile_id": "prf_b41d09",
      "match_score": 95,
      "match_status": "Unreviewed",
      "categories": ["Adverse Media"],
      "country": "NG",
      "listed_on": ["Negative News — Financial Crime"]
    }
  ]
}

Identity first. The Match Score (name 60% / date of birth 25% / country 15%, default threshold 93) decides whether the news record is really your customer. Below threshold, the hit is auto-classified False Positive — which, for adverse media specifically, is what makes the signal usable at all.

Then risk. Adverse media is a category, and category carries 50% of the Risk Score by default (with country risk at 30% and criminal record at 20%). So an adverse-media hit in a high-risk jurisdiction contributes more to the decision than the same hit in a low-risk one.

Review states. Each hit carries False Positive, Unreviewed, Confirmed Match, or Inconclusive — the audit trail an examiner expects when you onboard, or decline, a customer with negative coverage.

Price. $0.20 per check — adverse media is included in the standard AML screening, not billed separately.

Capability deep-dive: why "grounded" matters here

"Grounded" adverse media means two specific things in Didit's model. First, every hit is tied to a structured profile with its category and the underlying source it appeared on, so an analyst sees why the record surfaced rather than a raw article dump. Second, every hit is filtered through identity confidence before risk — the Match Score gate runs first, so the analyst's queue contains people the engine believes are actually the customer, not everyone who shares a name with a news subject. The combination is what turns adverse media from a liability (noise, alert fatigue, switched-off checks) into a genuine early-warning layer.

Use cases

• Fintech. Catch customers under public investigation before they're ever formally listed, as part of standard due diligence.
• Crypto / Web3. Pair adverse-media screening of the human with on-chain wallet screening for a complete counterparty risk view.
• Lending. Screen borrowers and guarantors for negative news on fraud or insolvency before disbursing.
• Marketplaces. Check high-value sellers for adverse coverage without drowning onboarding in namesake news stories.
• iGaming. Apply adverse-media checks to enhanced due diligence on high-value players and document each decision.

How to integrate with Didit

1. Add AML screening as a workflow step or call POST /v3/aml/ directly — adverse media is included.
2. Branch on the category. Route Adverse Media hits into your enhanced-due-diligence flow as your policy requires.
3. Tune identity confidence. Lean on the Match Score threshold (and document-number Golden Key) to keep namesake news out of review.
4. Enable ongoing monitoring so adverse coverage emerging after onboarding is flagged the next day.

Frequently asked questions

Is adverse media a separate product?

No. It's one category screened within the standard AML check against 1,300+ lists, included in the $0.20-per-check price.

How do you stop adverse media from flooding my analysts?

The Match Score gate runs first. A negative-news record that doesn't clear your identity threshold (default 93) is auto-classified False Positive, so analysts only see hits the engine believes are genuinely your customer.

How does adverse media affect the decision?

It's a category, and category carries 50% of the Risk Score by default. A confirmed adverse-media hit in a higher-risk jurisdiction contributes more than the same hit elsewhere.

Will I catch news that appears after onboarding?

Yes, if you enable ongoing monitoring — daily rescreening surfaces adverse media that emerges after a customer is onboarded.

What record do I keep for a regulator?

Every adverse-media hit retains its Match Score, Risk contribution, and review state (False Positive, Unreviewed, Confirmed Match, or Inconclusive) — a defensible audit trail for each decision.

Ready to get started?

Read the AML Screening overview in the docs, see adverse media in the full screening on the AML Screening product page, and check transparent per-check pricing on the pricing page. When you're ready, start free — 500 free KYC checks every month, with AML screening at $0.20 per check.