What Is AML Screening? A Complete Guide to Types, Process, and Best Practices for 2026

In 2025, global regulators imposed $3.8 billion in AML, KYC, sanctions, and CDD penalties — with crypto platforms, neobanks, and broker-dealers absorbing the sharpest hits. OKX paid $504 million. BitMEX paid $100 million. Canaccord Genuity was hit with an $80 million FinCEN penalty — the largest ever against a broker-dealer — for failures that included unfiled SARs and inadequate customer due diligence.

At the center of nearly every one of these enforcement actions is the same foundational failure: AML screening that wasn't working.

AML screening is the process of checking customers, entities, transactions, and payments against sanctions lists, politically exposed person (PEP) databases, adverse media sources, and global watchlists to identify and manage money laundering and terrorist financing risk. It is not a one-time onboarding check. It is a continuous, risk-based process that sits at the foundation of every compliance program — and when it fails, the consequences are measured in hundreds of millions of dollars.

This guide covers what AML screening is, the different types of screening, how the process works in practice, the technical and operational challenges that trip up compliance teams, what the latest regulations require, and how to build a screening program that actually performs in 2026.

What Is AML Screening and Why Does It Matter?

AML screening is the systematic process of checking individuals, businesses, transactions, and payments against databases of known or suspected risks — sanctions lists, PEP databases, law enforcement watchlists, adverse media sources, and other intelligence feeds — to identify parties associated with money laundering, terrorist financing, sanctions violations, or other financial crimes.

It is the first line of defense in any compliance program. Before a customer is onboarded, before a payment is processed, before a business relationship is established — screening determines whether the other party poses a regulatory or financial crime risk that needs to be managed, escalated, or blocked entirely.

How Does AML Screening Differ from Transaction Monitoring?

This is one of the most common sources of confusion in compliance. AML screening and transaction monitoring are related but fundamentally different processes.

AML screening asks: "Who is this person or entity, and are they on any risk lists?" Transaction monitoring asks: "Is this person or entity behaving suspiciously based on their transaction patterns?"

Both are essential. Neither replaces the other. For a detailed comparison of these approaches, see Signzy's analysis of transaction screening vs. transaction monitoring.

What Are the Different Types of AML Screening?

AML screening is not a single activity — it encompasses several distinct screening types, each targeting different risk categories and using different data sources. Understanding these distinctions is critical for building a comprehensive program.

Sanctions Screening

Sanctions screening checks customers, entities, and transactions against lists of individuals, organizations, and countries subject to economic sanctions imposed by governments and international bodies. These lists include:

• OFAC (Office of Foreign Assets Control) — United States
• UN Security Council consolidated sanctions list — Global
• EU consolidated sanctions list — European Union
• OFSI (Office of Financial Sanctions Implementation) — United Kingdom
• SEBI and RBI sanctions lists — India
• Local regulatory lists in 180+ jurisdictions

Sanctions screening is binary in its regulatory requirement: processing a transaction involving a sanctioned party is illegal, regardless of whether the institution knew about the designation. This makes sanctions screening the most consequential form of AML screening — and the one where failures carry the harshest penalties.

In 2025, OFAC enforcement actions exceeded prior years, with eight of fourteen actions targeting Russia-related sanctions violations. A UK bank was fined £160,000 after a sanctioned individual opened an account because a single spelling discrepancy evaded its screening controls — a stark illustration of why exact-match screening alone is insufficient.

For a detailed guide to sanctions screening implementation, see Signzy's sanctions screening AML guide.

PEP Screening

PEP screening identifies Politically Exposed Persons — individuals who hold or have held prominent public functions, along with their family members and close associates. PEPs are not inherently criminal, but their positions expose them to higher corruption risk, which regulatory frameworks require institutions to manage through enhanced due diligence.

PEPs are typically classified into three tiers:

Unlike sanctions lists, there is no single authoritative PEP database. Institutions must either compile their own PEP data or source it from specialized RegTech vendors — making data quality and coverage a critical differentiator between screening solutions.

For a deeper dive into PEP identification and management, see Signzy's guide on PEPs and sanction checks.

Adverse Media Screening

Adverse media screening — also called negative news screening — scans news sources, court records, regulatory filings, government publications, and increasingly social media for negative information about individuals or entities. Unlike sanctions and PEP screening, which check against structured databases, adverse media screening must process unstructured data — articles, press releases, social media posts, and legal filings — to identify risk signals.

Adverse media screening is particularly valuable for detecting:

• Individuals or entities under investigation but not yet sanctioned
• Fraud allegations, corruption charges, or criminal proceedings
• Environmental, social, and governance (ESG) risk signals
• Connections to organized crime or terrorism not yet captured in watchlists

The challenge is precision. Unstructured data produces a high volume of irrelevant results unless the screening system uses natural language processing (NLP) and sentiment analysis to distinguish genuinely adverse information from neutral mentions. A person named in a news article about financial crime is very different from a person named in the same article as a victim or an investigator.

Payment and Transaction Screening

Payment screening checks individual transactions — both incoming and outgoing — against risk lists before they are processed. It operates in real time (or near-real time) to intercept payments involving sanctioned parties, high-risk jurisdictions, or entities flagged for other compliance concerns.

Payment screening evaluates:

• Sender and receiver names against sanctions and watchlists
• Originator and beneficiary bank details
• Payment amounts and currencies against jurisdiction-specific thresholds
• Geographic risk based on transaction routing (country of origin, destination, intermediaries)
• Free-text payment fields for references to sanctioned entities or suspicious descriptors

The key distinction from customer screening is timing: payment screening must happen fast enough to not disrupt payment processing while still catching compliance risks. For high-volume payment processors, this creates a tension between screening thoroughness and straight-through processing rates.

Customer Screening and CDD Integration

Customer screening is the broadest form of AML screening, typically performed as part of Customer Due Diligence (CDD) during onboarding and at regular intervals throughout the business relationship. It combines sanctions screening, PEP screening, and adverse media screening into a single process that generates a risk profile for each customer.

The customer screening process typically follows these steps:

1. Data collection — Gathering identifying information (name, date of birth, address, nationality, ID documents; for businesses: registration details, UBO information)
2. Identity verification — Confirming the customer's identity through document verification, biometric checks, or database validation
3. Name screening — Running the customer's name and identifying details against sanctions lists, PEP databases, adverse media, and watchlists
4. Risk scoring — Assigning a risk level based on screening results, customer type, geography, product, and other factors
5. Decision — Low-risk customers proceed to onboarding; medium/high-risk customers require Enhanced Due Diligence (EDD); matches to sanctions lists result in blocking

For KYB (Know Your Business) screening, the process extends to verifying the business entity, tracing ownership structures to identify UBOs, and screening each UBO individually. For a detailed guide on business verification, see Signzy's blog on how to check if a company is legitimate.

Comparison of AML Screening Types

How Does the AML Screening Process Work?

While specific implementations vary by organization, industry, and jurisdiction, the AML screening process follows a consistent workflow — from data collection through resolution.

For a detailed process walkthrough with examples, see Signzy's AML watchlist screening guide.

What Is Fuzzy Matching and Why Is It Critical for AML Screening?

Fuzzy matching is the use of algorithms that identify approximate or partial matches between names and identifiers, rather than requiring exact character-for-character matches. It is arguably the most important technical capability in any AML screening system — because in the real world, names are never consistent.

Why Exact Matching Fails

Consider a sanctioned individual named Mohammed Al-Rahman. In different databases, documents, and transaction records, this name might appear as:

• Mohammad Al-Rahman
• Mohamed Alrahman
• Muhammed Al Rahman
• M. Al-Rahman
• محمد الرحمن (Arabic script)
• Мухаммед Аль-Рахман (Cyrillic transliteration)

An exact-match system would miss every one of these variations. And this is not a hypothetical problem — the UK bank fined £160,000 in 2025 lost a sanctioned individual precisely because a spelling discrepancy evaded its screening.

How Fuzzy Matching Works

Modern AML screening systems use multiple matching algorithms in combination:

Effective screening systems combine these algorithms and apply configurable thresholds that balance sensitivity (catching true matches) against specificity (minimizing false positives). The calibration of these thresholds is one of the most consequential technical decisions in any AML program.

What Are the Biggest Challenges in AML Screening?

The False Positive Problem

False positives are the single biggest operational challenge in AML screening. Industry data consistently shows that 85–95% of screening alerts are false positives — matches that, upon review, turn out to be legitimate customers with no connection to the flagged risk.

According to WorkFusion's analysis, false positives create significant backlogs that increase regulatory risk, as potential true matches remain undetected while analysts work through low-quality alerts. Flagright's compliance research found that compliance teams spend up to 90% of their effort on alerts that turn out to be non-actionable — effort that could be directed toward investigating genuine risks.

This is not just an efficiency problem. Gartner predicts that legal and compliance departments will increase investments in governance, risk, and compliance tools by 50% by 2026, driven in large part by alert fatigue and false positive management challenges.

Data Quality and Timeliness

Screening is only as good as the data feeding it — both the customer data being screened and the risk lists being screened against. Common data quality issues include:

• Incomplete customer records — Missing dates of birth, nationality, or address details that would help disambiguate matches
• Stale list data — Sanctions designations change daily; a system using weekly list updates creates a window of exposure
• Inconsistent data formats — Customer data collected in different formats across systems and channels
• Duplicate records — Multiple customer records for the same individual, each with slightly different data

Cross-Language and Cross-Script Matching

Global institutions must screen names that originate in Arabic, Chinese, Cyrillic, Devanagari, and dozens of other scripts. Transliteration from non-Latin scripts introduces enormous variation — the same Arabic name can be romanized in 20+ different ways. This challenge is compounded by cultural naming conventions that differ from Western norms (patronymics, tribal names, honorifics, single names).

Regulatory Complexity Across Jurisdictions

An institution operating in the US, EU, India, and the UAE must screen against different list sets, apply different risk thresholds, and comply with different reporting requirements in each jurisdiction. The EU's updated high-risk third country list (effective January 29, 2026) added Bolivia, the British Virgin Islands, and Russia — requiring institutions to apply enhanced due diligence and increased monitoring frequency for relationships connected to these jurisdictions.

Balancing Thoroughness with Customer Experience

Every additional screening step adds friction to the customer onboarding process. For fintechs and neobanks competing on speed and user experience, the challenge is implementing thorough screening without creating abandonment-inducing delays. This is where screening technology choices — real-time API-based screening vs. batch processing, intelligent risk-based routing — become critical business decisions, not just compliance decisions.

For a comprehensive overview of AML red flags that screening should detect, see Signzy's guide on AML red flags.

How Is AI Transforming AML Screening?

The application of artificial intelligence and machine learning to AML screening is not theoretical — it is the primary driver of operational improvement in compliance programs today.

Where AI Adds Value in AML Screening

The Regulatory Perspective on AI in Screening

Regulators are increasingly supportive of AI adoption in compliance — but with conditions. The key requirement is explainability: institutions must be able to explain to regulators why a specific screening decision was made, what data informed it, and how the model arrived at its conclusion. Black-box AI that cannot be audited is a regulatory liability, not an asset.

FATF's revised Recommendation 1 (February 2025) explicitly supports technology-enabled compliance, including digital onboarding and proportional due diligence — providing regulatory backing for AI-driven screening approaches that maintain appropriate risk controls.

Continuous Screening vs. Point-in-Time Screening: Why It Matters

One of the most significant shifts in AML screening practice is the move from periodic, point-in-time screening to continuous, real-time screening.

The regulatory direction is clear. FATF, the EU's AMLA framework, and FinCEN all increasingly expect institutions to demonstrate that their screening is not just periodic but responsive to changes in risk — including new sanctions designations, PEP status changes, and emerging adverse media. An institution that screens a customer at onboarding and doesn't rescreen for 12 months is creating a 364-day window in which that customer could be designated without the institution knowing.

What Are the Key Regulatory Frameworks Driving AML Screening?

AML screening requirements are not optional — they are driven by specific regulatory frameworks that vary by jurisdiction but share common principles.

For a comprehensive AML compliance checklist, see Signzy's guide on AML compliance pillars, red flags, and processes.

What Are the AML Screening Requirements by Industry?

Different industries face different screening requirements based on their risk profiles, customer types, and regulatory environments.

For fintechs specifically, the challenge is that screening infrastructure must scale alongside customer growth — and sponsor banks increasingly demand evidence of screening effectiveness as a condition of maintaining the BaaS relationship. For a comprehensive guide to AML policy for fintechs, see Signzy's AML policy for fintechs guide.

AML Screening Failures: What Enforcement Data Tells Us

The enforcement record from 2025–2026 provides a clear picture of what regulators expect from AML screening — and what failures look like.

Major AML Screening-Related Enforcement Actions (2025–2026)

What These Enforcement Actions Reveal

Three patterns emerge consistently across these cases:

1. Screening systems that weren't calibrated to the business. The Canaccord Genuity case is instructive — FinCEN found that the firm's AML surveillance system was not only understaffed but also produced reports that were never analyzed. Buying a screening tool is not the same as operating it effectively.

2. Growth that outpaced controls. Cash App's $40 million penalty and OKX's $504 million penalty both involved platforms that scaled rapidly without proportionally scaling their compliance infrastructure. As compliance industry commentary has noted: "Sanctions screening failures rarely stem from the technology itself" — they stem from how systems are "configured, governed, and operated."

3. Continuous monitoring gaps. Multiple enforcement actions cited failures in ongoing screening — not just onboarding screening. A customer who was clean at onboarding but subsequently designated remains the institution's risk if the institution doesn't rescreen.

For a deeper understanding of how money laundering works and where screening fits in the detection chain, see Signzy's guide on the three stages of money laundering.

How Signzy Helps Organizations Automate AML Screening

The operational challenges are clear: lean compliance teams, high false positive rates, cross-jurisdiction complexity, and screening systems that must keep pace with daily list changes while processing thousands of checks. Running AML screening across separate point solutions — one for sanctions, another for PEP, another for adverse media — creates workflow fragmentation, weak audit trails, and a higher total cost of ownership.

Signzy provides integrated AML screening infrastructure trusted by over 1,000 financial institutions globally:

• Sanctions and watchlist screening against 1,000+ global databases — including OFAC, UN, EU, FinCEN, SEBI, RBI, and local regulatory lists across 240+ countries — with daily list updates ensuring new designations are reflected within hours, not weeks.
• Fuzzy logic matching that catches name variations, aliases, transliterations, and misspellings that exact-match systems miss — the same type of discrepancy that cost a UK bank £160,000 when a single spelling variation evaded its screening.
• PEP screening across global databases covering Level 1–3 PEPs and their relatives and close associates, with continuous monitoring for status changes.
• Adverse media screening powered by AI to identify genuinely adverse information from unstructured sources while filtering noise.
• Real-time API-first architecture — 340+ REST API endpoints that integrate into existing onboarding, payment, and compliance workflows, deployable in 2–4 days with sub-5-second response times.
• No-code workflow builder that allows compliance teams to configure screening flows, adjust risk thresholds, and deploy screening rules without developer resources — critical when screening requirements change with every new regulation.
• Usage-based pricing with no minimum commitments — contrasting with enterprise-only pricing models at many competitors, enabling startups and scaling fintechs to adopt comprehensive screening without large upfront investments.

For organizations that need screening as part of a broader compliance stack — covering KYC, KYB, transaction monitoring, fraud detection, and bank account verification within a single platform — Signzy's unified infrastructure eliminates the vendor fragmentation that creates data silos and inconsistent risk scoring. For a detailed comparison of AML screening tools, see the top 10 AML watchlist screening tools for 2026.