Edition: computed from the 56 fintech profiles as of Nov 19, 2026; the rule is stated under Method. The bank equivalent: which banks lead in AI for fraud detection and credit decisioning?
By fraud, anti-money-laundering and identity AI in production and disclosed results, the fintechs leading in AI fraud prevention and risk are Visa, Monzo, Checkout.com, Plaid and Revolut: Visa (2 of 4 in production, led by A2A Protect (with Featurespace)); Monzo (3 of 3 in production, led by Fraud Prevent agent); Checkout.com (2 of 2 in production, led by Fraud Detection Pro); Plaid (2 of 2 in production, led by Trust Index (Plaid Protect)); Revolut (2 of 2 in production, led by AI card-scam detection). 33 of the 36 fintechs with a fraud or financial-crime AI use case on their page run one in production, and 15 disclose a result figure.
Data as of Nov 19, 2026 · 56 fintechs and neobanks profiled · 36 scored · 68 sources cited below · ranked answer, not a comparison
Which fintechs run AI for fraud, scams and financial crime, and what do they report?
The 30 highest of the 36 fintechs with a use case tagged fraud or AML/KYC on their page, each with its stage. Columns: the use cases, the latest disclosed result about fraud, scams, false positives, losses, alerts or reviews, the latest dated fraud or financial-crime AI move, and the score.
| # | Company | Fraud and financial-crime use cases (stage) | Disclosed result | Latest fraud or risk AI move | In production / score | Status | Date | Deeper | Sources |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Visa Card network | A2A Protect (with Featurespace) (in production) · Scam Disruption with generative AI tools (in production) · Large Transaction Model (announced) | Increase in fraud detection with A2A Protect, first six months: 75% (Sep 1, 2026) | SESA 2.0 and Visa Commercial Fraud Solutions announced (Oct 21, 2026) | 2 of 4 in production · score 10 | In production | Oct 21, 2026 | Visa: fraud and risk AI | Visa ↗ · Visa ↗ · Visa ↗ |
| 2 | Monzo Neobank | Fraud Prevent agent (in production) · Representation learning fraud models (in production) · Payment Limit Increase Agent and Fraud Protect agents (in production) | Lift in authorised fraud prevented by Fraud Prevent agent: Over 20% (Jul 27, 2026) | Fraud Prevent agent live in the payment path (Jul 27, 2026) | 3 of 3 in production · score 8 | In production | Jul 27, 2026 | profile | Monzo ↗ · Monzo ↗ · Monzo ↗ |
| 3 | Checkout.com Payments | Fraud Detection Pro (in production) · AI-driven policy and due-diligence reviews (in production) | Due-diligence time cut by AI policy reviews: 83% (Feb 24, 2026) | Fraud Detection Pro launches (Oct 19, 2022) | 2 of 2 in production · score 8 | In production | Feb 24, 2026 | profile | Checkout.com ↗ · Checkout.com ↗ |
| 4 | Plaid Business finance & infrastructure | Trust Index (Plaid Protect) (in production) · LLM document checks in Identity Verification (in production) | Trust Index 3 fraud detection: up to 41% more fraud caught at the same false-positive rate (testing) (May 21, 2026) | ML risk scores, LendScore beta and a new CFO (Oct 20, 2025) | 2 of 2 in production · score 8 | In production | Oct 6, 2026 | Plaid: fraud and risk AI | Plaid ↗ · Plaid ↗ · Plaid ↗ |
| 5 | Revolut Neobank | AI card-scam detection (in production) · AI-assisted financial-crime review (in production) | PRAGMA early performance gains vs. legacy systems: 2.3x credit-risk accuracy; 65% more fraud caught; 41% better recommendations (Aug 25, 2026) | AI financial-crime detection outperforms manual review, US CEO says (May 27, 2026) | 2 of 2 in production · score 8 | In production | Aug 25, 2026 | profile | Revolut ↗ · Revolut ↗ · PYMNTS.com ↗ |
| 6 | Stripe Payments | Radar and the Payments Foundation Model (in production) · Radar for token theft and sign-up abuse (in production) | Radar, AI sign-up fraud: 1 in 6 attempted sign-ups at AI services on Stripe made by a bad actor; more than 3.3 million risky sign-ups blocked in a month for eight AI businesses (Apr 29, 2026) | no dated fraud or risk AI move | 2 of 2 in production · score 8 | In production | May 7, 2026 | Stripe: fraud and risk AI | Stripe ↗ · Stripe ↗ · Fortune ↗ |
| 7 | Mastercard Card network | Large tabular foundation model (pilot) · Agentic transaction probability score (pilot) · Decision Intelligence Pro (announced) | Reduction in false positives (Mastercard analysis): More than 85% (Feb 1, 2024) | no dated fraud or risk AI move | 0 of 3 in production · score 7 | Pilot | Sep 30, 2026 | Mastercard: fraud and risk AI | Mastercard ↗ · Mastercard ↗ · Mastercard ↗ |
| 8 | Starling Neobank | Scam Intelligence (in production) · Invoice-fraud checks in Starling Assistant (in production) · Agentic financial-crime investigator (in production) | no result disclosed | Agentic financial-crime investigator embedded in the core (Oct 21, 2026) | 3 of 3 in production · score 6 | In production | Oct 21, 2026 | profile | Starling ↗ · Starling ↗ · Finextra ↗ |
| 9 | Brex Business finance & infrastructure | Audit agent and fraud simulation testing (in production) · Multi-agent customer onboarding (in production) | Onboarding agents: card-application auto-approval 0% to 40%; manual identity reviews down 70%; business-address RFIs down 85% (Jan 20, 2026) | no dated fraud or risk AI move | 2 of 2 in production · score 6 | In production | Apr 8, 2026 | profile | American Banker ↗ · Brex ↗ · Brex ↗ |
| 10 | Chime Neobank | AI fraud detection and risk scoring (in production) · AI disputes management platform (in production) | Fraud-loss reduction since 2022 (AI fraud and risk scoring): 29% (Jan 14, 2026) | Transformer risk models disclosed (Aug 27, 2026) | 2 of 2 in production · score 6 | In production | Aug 27, 2026 | profile | Chime ↗ · Chime ↗ |
| 11 | MariBank Neobank | AI income-document verification (Monee) (in production) · AI fraud detection plus in-app scam controls (in production) | Monee income-document review time: reduced by around 95% with AI verification (Jun 30, 2026) | Sea adopts NVIDIA Vera Rubin; Monee AI in fraud and credit (Sep 23, 2026) | 2 of 2 in production · score 6 | In production | Oct 24, 2026 | profile | The Motley Fool (earnings call transcript) ↗ · Sea Limited ↗ · MariBank ↗ |
| 12 | N26 Neobank | Financial-crime analysis support (in production) · Real-time transaction fraud detection (in production) | Fraud volume reduction (AWS case study): 80% in 2023; decisions within 500 ms (Dec 31, 2023) | Real-time fraud engine goes into production (Jun 1, 2023) | 2 of 2 in production · score 6 | In production | Oct 8, 2026 | profile | Anthropic ↗ · N26 ↗ · Amazon Web Services ↗ |
| 13 | Remitly Payments | ML fraud scoring and the 2025 AI fraud model (in production) · Electronic KYC and real-time payment authentication (in production) | Q4 2025 transaction losses: record low 7.3 basis points of send volume; about $10 million incremental RLTE from the upgraded AI fraud model (Dec 31, 2025) | Q4 2025 call: AI fraud model drives record-low losses (Feb 18, 2026) | 2 of 2 in production · score 6 | In production | Mar 13, 2026 | profile | Remitly ↗ · earnings call transcript (The Motley Fool) ↗ · Remitly via SEC EDGAR ↗ |
| 14 | Robinhood Brokerage | ML fraud detection and support automation (in production) · FinCrimes Agent (in production) | FinCrimes Agent efficiency gain (vendor-reported): roughly 20% cumulative efficiency gain in investigative workflows (Oct 16, 2026) | AWS case study: generative-AI FinCrimes Agent (Oct 16, 2026) | 2 of 2 in production · score 6 | In production | Oct 16, 2026 | profile | U.S. Securities and Exchange Commission (EDGAR) ↗ · Amazon Web Services ↗ |
| 15 | Wise Payments | Machine-learning compliance checks (in production) · Real-time risk monitoring (in production) | Compliance checks completed in under one second by ML: 99% (Mar 31, 2026) | no dated fraud or risk AI move | 2 of 2 in production · score 6 | In production | Jun 25, 2026 | profile | Wise Group plc via SEC EDGAR ↗ · Wise Group plc via SEC EDGAR ↗ |
| 16 | Klarna BNPL & lending | Real-time fraud detection (in production) · AML and sanctions screening (in production) | no result disclosed | no dated fraud or risk AI move | 2 of 2 in production · score 4 | In production | Sep 10, 2025 | profile | Klarna Group plc via SEC EDGAR ↗ · Finansinspektionen ↗ |
| 17 | Upstart BNPL & lending | Verification and fraud-detection models (in production) · Automated identity verification (in production) | no result disclosed | no dated fraud or risk AI move | 2 of 2 in production · score 4 | In production | Feb 10, 2026 | profile | Upstart Holdings via SEC EDGAR ↗ |
| 18 | Coinbase Brokerage | LLM risk agent for Onramp (in production) | Onramp fraud agent: 30% fewer fraudulent transactions, 22% less fraud value (A/B test); post-trained model 55% lower latency than Opus 4.5 (Oct 8, 2026) | Engineering series: a post-trained fraud agent beats Opus 4.5 (Oct 8, 2026) | 1 of 1 in production · score 4 | In production | Oct 8, 2026 | profile | Coinbase ↗ |
| 19 | Adyen Payments | Uplift Protect (risk module) (in production) · Dynamic Identification for onboarding and compliance (announced) | no result disclosed | no dated fraud or risk AI move | 1 of 2 in production · score 3 | In production | Sep 3, 2026 | Adyen: fraud and risk AI | Adyen ↗ · Adyen ↗ · Adyen ↗ |
| 20 | Circle Payments | AI-assisted AML alert review (in production) · Agent identity verification with Proof (x401) (announced) | no result disclosed | no dated fraud or risk AI move | 1 of 2 in production · score 3 | In production | Oct 19, 2026 | profile | Circle Internet Group via SEC EDGAR ↗ · Circle ↗ · PYMNTS ↗ |
| 21 | Payoneer Business finance & infrastructure | AI recipient-name matching (in production) · Oscilar AI risk decisioning (rolling out) | no result disclosed | Oscilar selected for AI fraud and risk decisioning (Dec 11, 2025) | 1 of 2 in production · score 3 | In production | Dec 11, 2025 | profile | Payoneer Engineering ↗ · Oscilar via PR Newswire ↗ |
| 22 | Affirm BNPL & lending | ML fraud detection (in production) | no result disclosed | no dated fraud or risk AI move | 1 of 1 in production · score 2 | In production | Aug 27, 2026 | profile | Affirm Holdings via SEC EDGAR ↗ · Affirm Holdings via SEC EDGAR ↗ |
| 23 | Block Payments | Machine-learning risk assessment (in production) | no result disclosed | no dated fraud or risk AI move | 1 of 1 in production · score 2 | In production | Aug 5, 2026 | profile | Block, Inc. via SEC EDGAR ↗ · Block, Inc. via SEC EDGAR ↗ |
| 24 | Bluevine Neobank | ML in check processing and fraud controls (in production) | no result disclosed | no dated fraud or risk AI move | 1 of 1 in production · score 2 | In production | Sep 28, 2026 | profile | Forbes ↗ · Valley National Bancorp ↗ |
| 25 | bunq Neobank | AI-trained fraud and deepfake detection (in production) | no result disclosed | no dated fraud or risk AI move | 1 of 1 in production · score 2 | In production | Jan 7, 2026 | profile | bunq ↗ · bunq ↗ |
| 26 | Dave Neobank | AI fraud mitigation (in production) | no result disclosed | no dated fraud or risk AI move | 1 of 1 in production · score 2 | In production | Mar 2, 2026 | profile | Dave Inc. via SEC EDGAR ↗ |
| 27 | Marqeta Payments | Real-Time Decisioning ML Risk Score (in production) | no result disclosed | Forter joins the RTD AI-powered Risk Score (Nov 5, 2026) | 1 of 1 in production · score 2 | In production | Nov 5, 2026 | profile | Marqeta via Business Wire ↗ · Marqeta, Inc. via SEC EDGAR ↗ · Forter via PR Newswire ↗ |
| 28 | Nubank Neobank | AI in risk and fraud-loss reduction (in production) | no result disclosed | no dated fraud or risk AI move | 1 of 1 in production · score 2 | In production | Nov 13, 2025 | profile | Nu ↗ |
| 29 | OakNorth Neobank | AI and automated tools for fraud and financial-crime screening (in production) | no result disclosed | no dated fraud or risk AI move | 1 of 1 in production · score 2 | In production | Jul 25, 2025 | profile | OakNorth ↗ |
| 30 | Oportun BNPL & lending | Fraud models (in production) | no result disclosed | no dated fraud or risk AI move | 1 of 1 in production · score 2 | In production | Feb 27, 2026 | profile | Oportun via SEC EDGAR ↗ |
Each row quotes the company's page on this site; the company link opens the full record with every source. Companies below the 30 shown are scored but not listed; draft profiles are not counted.
Which fintechs disclose a result from their fraud or risk AI?
Every company in the ranking with a disclosed result figure, the latest one on its page. The figures use different bases and are not comparable with each other.
| Company | Category | Latest disclosed result | As of |
|---|---|---|---|
| Visa | Card network | Increase in fraud detection with A2A Protect, first six months: 75% | Sep 1, 2026 |
| Monzo | Neobank | Lift in authorised fraud prevented by Fraud Prevent agent: Over 20% | Jul 27, 2026 |
| Checkout.com | Payments | Due-diligence time cut by AI policy reviews: 83% | Feb 24, 2026 |
| Plaid | Business finance & infrastructure | Trust Index 3 fraud detection: up to 41% more fraud caught at the same false-positive rate (testing) | May 21, 2026 |
| Revolut | Neobank | PRAGMA early performance gains vs. legacy systems: 2.3x credit-risk accuracy; 65% more fraud caught; 41% better recommendations | Aug 25, 2026 |
| Stripe | Payments | Radar, AI sign-up fraud: 1 in 6 attempted sign-ups at AI services on Stripe made by a bad actor; more than 3.3 million risky sign-ups blocked in a month for eight AI businesses | Apr 29, 2026 |
| Mastercard | Card network | Reduction in false positives (Mastercard analysis): More than 85% | Feb 1, 2024 |
| Brex | Business finance & infrastructure | Onboarding agents: card-application auto-approval 0% to 40%; manual identity reviews down 70%; business-address RFIs down 85% | Jan 20, 2026 |
| Chime | Neobank | Fraud-loss reduction since 2022 (AI fraud and risk scoring): 29% | Jan 14, 2026 |
| MariBank | Neobank | Monee income-document review time: reduced by around 95% with AI verification | Jun 30, 2026 |
| N26 | Neobank | Fraud volume reduction (AWS case study): 80% in 2023; decisions within 500 ms | Dec 31, 2023 |
| Remitly | Payments | Q4 2025 transaction losses: record low 7.3 basis points of send volume; about $10 million incremental RLTE from the upgraded AI fraud model | Dec 31, 2025 |
| Robinhood | Brokerage | FinCrimes Agent efficiency gain (vendor-reported): roughly 20% cumulative efficiency gain in investigative workflows | Oct 16, 2026 |
| Wise | Payments | Compliance checks completed in under one second by ML: 99% | Mar 31, 2026 |
| Coinbase | Brokerage | Onramp fraud agent: 30% fewer fraudulent transactions, 22% less fraud value (A/B test); post-trained model 55% lower latency than Opus 4.5 | Oct 8, 2026 |
How is this ranking computed?
Data: the 56 published fintech profiles, as of Nov 19, 2026. Use cases are the ones each page tags fraud or AML/KYC (the tag is set when the page is written and checked by the validator), kept when their text names AI, a model, machine learning, an agent, an assistant or automation; a use case whose text says it is not AI, such as a manual savings lock, or that describes a control without saying how it works, is left out. A result is a disclosed number about fraud, scams, false positives or declines, chargebacks, losses avoided, detection, alerts, due diligence, identity reviews or investigations; fines, penalties, settlements and loss allowances are not results and are left out.
Score = 2 × fraud and AML/KYC use cases in production + 1 × those at other stages + 2 × disclosed results (capped at 2). Ties: use cases in production, then results, then name. 36 companies qualify; the top 30 are shown.
What are the caveats?
Fraud AI is close to universal among fintechs that move money, and many describe it in a sentence of a filing: a company that names one model scores one use case, a company that publishes engineering posts about four scores four. Results are self-reported, often from a test or a pilot, on bases that differ.
Enforcement history is not scored here: several companies in the table have AML penalties on their pages, which their profiles record separately. A company missing from the table has no qualifying entry on its page on this site; that is a gap in its public record as recorded here, not proof that it does none of this work.
The bank ranking on the nearest question, which banks lead in AI for fraud detection and credit decisioning?, is computed from the bank profiles with its own rule; the peer groups and their sources differ, so positions are not comparable across the two pages.
Which fintech leads in AI fraud prevention?
Visa: 2 of 4 fraud and financial-crime AI use cases in production, led by A2A Protect (with Featurespace) and Scam Disruption with generative AI tools; its page reports Increase in fraud detection with A2A Protect, first six months: 75% (Sep 1, 2026).
Which fintechs disclose results from fraud AI?
Visa (Increase in fraud detection with A2A Protect, first six months: 75%), Monzo (Lift in authorised fraud prevented by Fraud Prevent agent: Over 20%), Checkout.com (Due-diligence time cut by AI policy reviews: 83%), Plaid (Trust Index 3 fraud detection: up to 41% more fraud caught at the same false-positive…), Revolut (PRAGMA early performance gains vs. legacy systems: 2.3x credit-risk accuracy; 65% more…) and Stripe (Radar, AI sign-up fraud: 1 in 6 attempted sign-ups at AI services on Stripe made by a…). The second table lists all 15.
Which fintechs use AI for anti-money laundering?
15 have an AML or KYC AI use case in production: Monzo (Payment Limit Increase Agent and Fraud Protect agents), Checkout.com (AI-driven policy and due-diligence reviews), Plaid (LLM document checks in Identity Verification), Revolut (AI-assisted financial-crime review), Brex (Multi-agent customer onboarding), MariBank (AI income-document verification (Monee)), N26 (Financial-crime analysis support) and Remitly (Electronic KYC and real-time payment authentication).
How many fintechs use AI for fraud detection?
33 of the 56 profiled fintechs describe fraud or financial-crime AI in production; 36 describe it at some stage.
How does this relate to the bank fraud and credit ranking?
The bank ranking at /banks/leaders/fraud-and-credit-ai scores fraud, AML and credit AI together for the large banks; this page takes fraud and financial crime alone for fintechs, and the credit-decisioning page covers fintech credit models.
When any of these companies moves on AI, you'll read it the next morning.
when a bank, fintech or regulator moves on AI, it is in the next morning's brief · six sourced stories · 7 am PT · free
plus every regulation, bank and agent page update, the morning after · leave any morning