# Which banks lead in AI for fraud detection and credit decisioning?

Source: https://www.bankingnewsai.com/banks/leaders/fraud-and-credit-ai
Last updated: Oct 15, 2026


Data as of Oct 6, 2026 · 140 banks profiled · 131 with evidence · 85 sources cited.

By fraud, AML and credit AI use cases in production, the banks that lead are Zions, Citi, Western Alliance, FNBO and SoFi: Zions (6 in production of 9; Fraud-related loan loss, Q3 2025: $50 million (Oct 20, 2025)); Citi (7 in production of 7; Cards approval-rate gain from new AI models: +100 basis points (May…); Western Alliance (6 in production of 6; Fraudulent class-action claims prevented in 2024: 723 million+ (Apr…); FNBO (5 in production of 7; Fraud-case investigation with Pindrop Fraud Assist…); SoFi (6 in production of 7; Quality reached by Poetic's AI on end-to-end fraud investigations…). 104 of the 140 profiled banks describe AI for fraud detection or AML in production and 57 describe AI in credit decisioning in production; fraud is the most widely disclosed AI use in banking, credit the most regulated.

## Which banks run AI for fraud, AML and credit decisioning in production?

The 30 highest of 131 banks with a fraud, AML/KYC, credit-underwriting or fair-lending AI use case on their page, each with its stage. Figure: the first disclosed number on the bank's page about fraud, AML, approvals, losses or alerts (a result where the bank gives one, a loss or exposure figure where that is what it reports); "results not disclosed" means the bank describes the system but no figure.

| # | Bank | Fraud and AML use cases (stage) | Credit decisioning use cases (stage) | Disclosed figure (fraud, AML or credit) | In production / score | Status | Date | Sources |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | [Zions](https://www.bankingnewsai.com/banks/zions) | AI in targeted fraud-detection workflows (in production) · AI for client authentication (in production) · AI on wire-room and AML/BSA false positives (in production) | Unstructured-document processing for onboarding and origination (in production) · AI in credit review, appraisal review and financial-statement spreading (in production) · AI in credit examination and treasury-management operations (rolling out) | Fraud-related loan loss, Q3 2025: $50 million (Oct 20, 2025) | 6 in production of 9 · score 17 | In production | Aug 5, 2026 | [American Banker](https://americanbanker.com/news/zions-revamped-tech-leadership-team-eyes-ai-stablecoins); [Yahoo Finance](https://finance.yahoo.com/news/zions-zion-q4-2025-earnings-000408635.html); [Earnings call transcript (The Motley Fool)](https://fool.com/earnings/call-transcripts/2025/08/05/zions-zion-q2-2025-earnings-call-transcript) |
| 2 | [Citi](https://www.bankingnewsai.com/banks/citigroup) | Faster account opening and onboarding (in production) · Behavioural biometrics (BioCatch) (in production) · Socure identity verification (in production) | Cards approval models (in production) | Cards approval-rate gain from new AI models: +100 basis points (May 7, 2026) | 7 in production of 7 · score 16 | In production | Sep 28, 2026 | [Reuters](https://www.reuters.com/business/finance/citigroup-says-ai-helps-speed-account-openings-systems-upgrades-2026-04-08/); [The Wall Street Journal](https://www.wsj.com/cio-journal/the-ai-superfans-companies-count-on-to-convert-the-skeptics-5b301a90); [American Banker](https://www.americanbanker.com/news/agentic-ai-shopping-bots-are-coming-banks-need-to-be-ready) |
| 3 | [Western Alliance](https://www.bankingnewsai.com/banks/western-alliance) | Triangle of Fraud Protection (Digital Disbursements) (in production) · AI triage in financial-crimes compliance (in production) · ClaimScore AI claim validation inside the Triangle (in production) | Credit agreement review and lending lifecycle redesign (in production) · AI as an underwriting theme in innovation banking (in production) · Venture debt to AI-product companies (Ordway) (in production) | Fraudulent class-action claims prevented in 2024: 723 million+ (Apr 4, 2025) | 6 in production of 6 · score 16 | In production | Aug 5, 2026 | [American Banker](https://digital-banking.americanbanker.com/sponsor/western-alliance); [Western Alliance Bank](https://westernalliancebancorporation.com/sites/default/files/2025-04/2025-digital-payments-research-report.pdf); [Western Alliance Bancorporation (SEC filings)](https://www.sec.gov/Archives/edgar/data/1212545/000162828026033850/wal_investordayx2026xfin.htm) |
| 4 | [FNBO](https://www.bankingnewsai.com/banks/fnbo) | Agentic alert triage and enhanced due diligence (in production) · Sanctions-screening agent (in production) · Pindrop Fraud Assist for card-fraud investigations (in production) | Upstart AI personal loans (nationwide expansion) (rolling out) · Upstart AI consumer-lending pilot (pilot) | Fraud-case investigation with Pindrop Fraud Assist (vendor-reported): Time −35–40%; analyst accuracy +50%; case productivity +42% (Mar 17, 2026) | 5 in production of 7 · score 16 | In production | Jun 24, 2026 | [American Banker](https://americanbanker.com/news/how-fnbo-uses-agentic-ai-to-investigate-financial-crime); [Pindrop (GlobeNewswire)](https://www.globenewswire.com/news-release/2026/03/17/3257231/0/en/pindrop-unveils-first-agentic-fraud-investigation-solution-to-combat-surging-ai-driven-fraud.html); [FNBO](https://www.fnbo.com/insights/fraud-and-security-tips/2025/ai-powered-scams-target-consumers) |
| 5 | [SoFi](https://www.bankingnewsai.com/banks/sofi) | AI fraud investigations (Poetic) (in production) · AI/ML real-time payment-risk platform for clients (in production) · Socure identity verification and fraud prevention (in production) | Personal Loan Doc Coach (in production) · Oscilar AI risk decisioning for credit strategy (in production) · Fairness-constrained ML training (patent) (announced) | Quality reached by Poetic's AI on end-to-end fraud investigations: 99%+ within five weeks (Jun 10, 2026) | 6 in production of 7 · score 15 | In production | Sep 24, 2026 | [PR Newswire (Poetic)](https://www.prnewswire.com/news-releases/poetic-raises-50m-series-a-to-automate-the-worlds-most-complex-enterprise-processes-with-reliable-ai-302796939.html); [SEC (SoFi Technologies 10-Q)](https://www.sec.gov/Archives/edgar/data/1818874/000181887426000054/sofi-20260630.htm); [Socure (Business Wire)](https://www.businesswire.com/news/home/20211109006037/en/Socure-Accelerates-Mission-to-be-the-First-to-Verify-100-of-Identities-and-Eliminate-Identity-Fraud-Across-all-Industries-with-a-%24450M-Investment-Led-by-Accel-and-T.-Rowe-Price-at-a-%244.5B-Valuation) |
| 6 | [Commonwealth Bank](https://www.bankingnewsai.com/banks/commonwealth-bank) | Agentic fraud-pattern detection and rule generation (in production) · Scam and fraud detection across payments (in production) · AI bots that engage and disrupt scammers (Apate.ai) (in production) | AI-assisted and agentic business lending (pilot) | Signals monitored daily by fraud protection systems: more than 80 million (Apr 24, 2026) | 5 in production of 6 · score 15 | In production | Oct 14, 2026 | [Commonwealth Bank of Australia](https://commbank.com.au/articles/newsroom/2026/04/ai-agent-spots-fraud-in-real-time.html); [Finextra](https://finextra.com/newsarticle/47683/commbank-builds-ai-agent-that-spots-fraud-and-helps-build-defences); [Commonwealth Bank of Australia](https://commbank.com.au/articles/newsroom/2026/02/cba-approach-to-adopting-ai-report-announcement.html) |
| 7 | [KeyBank](https://www.bankingnewsai.com/banks/keybank) | NiCE Actimize X-Sight AI platform for fraud and AML (in production) · AI-deepfake fraud defence and client education (in production) · Generative-AI anti-money-laundering screening (pilot) | AI research support for commercial loan approvals (in production) · AI to accelerate credit decisioning (announced) | AML screening time with generative AI: 15 to 20 minutes, against several hours manually (Aug 14, 2025) | 3 in production of 6 · score 13 | In production | Apr 16, 2026 | [NiCE Actimize (Business Wire)](https://www.businesswire.com/news/home/20250612909609/en/NiCE-Actimize-X-Sight-AI-Enterprise-Platform-Increases-Modernization-of-KeyBanks-Financial-Crime-Operations); [KeyBank](https://www.key.com/content/dam/kco/documents/businesses___institutions/payment-fraud-schemes.pdf); [Bank Automation News](https://bankautomationnews.com/allposts/ai/keybank-identifies-40-ai-proofs-of-concept) |
| 8 | [Synchrony](https://www.bankingnewsai.com/banks/synchrony) | Behavioural fraud detection and AI-enabled fraud defence (in production) · Cashflow analysis to identify credit washing (in production) | PRISM credit decisioning with cashflow data (in production) · PRISM across collections, identity and account management (in production) | Customers approved in 2025 via cashflow underwriting who would otherwise have been declined: Nearly 50,000 (up to 150,000 expected in 2026) (Dec 31, 2025) | 4 in production of 4 · score 12 | In production | Sep 30, 2026 | [PYMNTS](https://pymnts.com/data/2026/synchrony-puts-credit-in-the-test-kitchen); [Synchrony Financial (SEC filing)](https://www.sec.gov/Archives/edgar/data/1601712/000160171226000006/syf-20251231.htm); [Investing.com](https://ng.investing.com/news/transcripts/earnings-call-transcript-synchrony-beats-q2-2026-eps-forecast-as-sales-hit-record-93CH-2609405) |
| 9 | [BMO](https://www.bankingnewsai.com/banks/bmo-us) | AI alert triage and adverse-media search (in production) · FIS financial-crimes agent (Anthropic) (announced) | SmartDecision insurance underwriting (in production) · ARM automated review triage and portfolio monitoring (in production) · Aura commercial credit agent (rolling out) | Underwriting speed target: 50% faster by 2028 (Aug 6, 2026) | 3 in production of 5 · score 12 | In production | Aug 6, 2026 | [Yahoo Finance Canada](https://finance.yahoo.com/news/bmo-aims-increase-business-u-200703810.html); [Business Wire via FT Markets](https://markets.ft.com/data/announce/detail?dockey=600-202605041800BIZWIRE_USPRX____20260504_BW126906-1); [CNW via FT Markets](https://markets.ft.com/data/announce/detail?dockey=600-202607090730CANADANWCANADAPR_C9043-1) |
| 10 | [TD](https://www.bankingnewsai.com/banks/td-bank-us) | Fraud defence and claims (in production) · AI models in US AML monitoring and investigations (in production) · ML in transaction monitoring and the new KYC platform (in production) | Agentic lending pre-adjudication (rolling out) · Agentic mortgage funding and discharge (rolling out) | US BSA/AML remediation expense: $507 million fiscal 2025; about $550 million guided for fiscal 2026 (raised from $500 million) (Sep 17, 2026) | 3 in production of 6 · score 11 | In production | Oct 13, 2026 | [Banking Dive](https://www.bankingdive.com/news/td-ai-banking-customer-trust-jobs-virtual-assistant/756261/); [Banking Dive](https://www.bankingdive.com/news/td-bank-150m-insurance-cost-reductions-ai-chun-salom/813637/); [Motley Fool (Yahoo Finance)](https://finance.yahoo.com/markets/stocks/articles/td-td-q2-2026-earnings-184201327.html) |
| 11 | [First Hawaiian](https://www.bankingnewsai.com/banks/first-hawaiian) | Deepfake and scam awareness for customers (in production) · AI voice-cloning scam education (in production) · Planned Zest Protect fraud detection (announced) | AI-based automated underwriting in consumer lending (Zest AI) (in production) | Automated credit-card decisions (vendor-reported, Zest AI): 55% of applications, from 4%; instant approvals 40%, from 4% (Jun 5, 2025) | 3 in production of 4 · score 11 | In production | Feb 11, 2026 | [First Hawaiian Bank](https://www.fhb.com/en/resource-center/security-and-fraud-protection/deepfake-awareness-protecting-yourself-at-work-and-home); [First Hawaiian Bank](https://www.fhb.com/en/resource-center/security-and-fraud-protection/defend-yourself-against-ai-voice-scams); [Zest AI (vendor case study)](https://www.zest.ai/learn/success_stories/first-hawaiian-bank/) |
| 12 | [Scotiabank](https://www.bankingnewsai.com/banks/scotiabank) | Machine learning for fraud and cyber detection (in production) · Customer-record validation and AML detection on Scotia Intelligence (in production) · Tangerine AML AI pilot (pilot) | Credit-card pre-approval and mortgage verification (in production) | Tangerine AML pilot: reduction in existing alert volumes: 37% (Feb 24, 2026) | 3 in production of 4 · score 11 | In production | Sep 14, 2026 | [Bank Automation News](https://bankautomationnews.com/allposts/risk-security/scotia-explores-google-cloud-for-cybersecurity-solutions); [Scotiabank (CNW via Yahoo Finance)](https://finance.yahoo.com/news/scotiabank-digital-banking-latest-enhancements-110000432.html); [Scotiabank (The Bank of Nova Scotia)](https://www.scotiabank.com/content/dam/scotiabank/corporate/quarterly-reports/2026/q3/Q326_Investor_Presentation.pdf) |
| 13 | [Société Générale](https://www.bankingnewsai.com/banks/societe-generale) | Mosaic external-fraud detection (in production) · Instant-KYC remote onboarding (in production) · AI document handling in securities-services KYC (in production) | Nacre overdraft-overrun decisions and Stella credit analysis (in production) · Active Shell financial-fragility early warning (in production) | results not disclosed | 5 in production of 5 · score 10 | In production | Feb 1, 2025 | [Societe Generale](https://societegenerale.com/sites/default/files/documents/2024-09/applying-data-ia-flagship-use-cases.pdf); [Societe Generale Securities Services](https://securities-services.societegenerale.com/en/insights/views/news/post-trade-finds-its-feet-with-ai) |
| 14 | [Merchants Bank of Indiana](https://www.bankingnewsai.com/banks/merchants-indiana) | Wire-fraud customer education (in production) · Mortgage-fraud exposure in multifamily lending (in production) | Multifamily and healthcare mortgage banking with securitisation (in production) · Credit risk transfer on multifamily and healthcare pools (in production) | Q2 2025 provision increase and charge-offs: Provision up $43.1 million (432%) to $53.0 million; $46.1 million of charge-offs on 14 customers, linked to appraisals and mortgage fraud (Jun 30, 2025) | 4 in production of 4 · score 10 | In production | Jul 28, 2026 | [Merchants Bank of Indiana](https://merchantsbankofindiana.com/what-to-know-before-you-wire-money); [Merchants Bancorp (PR Newswire)](https://prnewswire.com/news-releases/merchants-bancorp-reports-second-quarter-2025-results-302515176.html); [Merchants Bancorp (PR Newswire)](https://prnewswire.com/news-releases/merchants-bancorp-reports-first-quarter-2026-results-302755998.html) |
| 15 | [Bread Financial](https://www.bankingnewsai.com/banks/bread-financial) | AI-enabled fraud and cyber threats (in production) · Proprietary fraud models including machine learning (in production) | Machine-learning credit models and possible deep learning (in production) · Collection scoring from 30 days past due (in production) | Fraud-related operational losses: $65 million in each of 2025 and 2024 (Dec 31, 2025) | 4 in production of 4 · score 10 | In production | Feb 13, 2026 | [Bread Financial Holdings (SEC filing)](https://www.sec.gov/Archives/edgar/data/1101215/000110121526000016/bfh-20251231.htm); [Investing.com](https://investing.com/news/transcripts/earnings-call-transcript-bread-financial-q4-2025-beats-eps-expectations-93CH-4481011); [Investing.com](https://investing.com/news/transcripts/earnings-call-transcript-bread-financial-q3-2025-reports-strong-earnings-beat-93CH-4305227) |
| 16 | [BofA](https://www.bankingnewsai.com/banks/bank-of-america) | AI fraud-detection models (in production) · Behavioural biometrics (BioCatch) (in production) · Person-to-person payment fraud analysis (in production) | no credit use case described | Fraud losses since 2018: cut in half through AI-enabled processes (Nov 5, 2025) | 3 in production of 3 · score 10 | In production | Jul 8, 2026 | [Bank of America (SEC filings)](https://www.sec.gov/Archives/edgar/data/70858/000007085825000254/bac063025ex992.htm); [American Banker](https://www.americanbanker.com/news/agentic-ai-shopping-bots-are-coming-banks-need-to-be-ready); [American Banker](https://americanbanker.com/news/revenue-opportunities-everywhere-bofas-tech-chief-on-ai) |
| 17 | [BNP Paribas](https://www.bankingnewsai.com/banks/bnp-paribas) | Real-time transaction fraud detection (in production) · Identity-theft and forged-document detection at Personal Finance (in production) · KYC and compliance with Mistral AI models (rolling out) | Document AI for French mortgage files (in production) · Conversational income-and-expenditure assessment (rolling out) · AI agents for corporate credit memos (announced) | results not disclosed | 3 in production of 7 · score 10 | In production | Oct 5, 2026 | [BNP Paribas](https://group.bnpparibas/en/group/about-us/frequently-asked-questions/faq-tech-artificial-intelligence); [BNP Paribas Personal Finance](https://personal-finance.bnpparibas/en/intelligence-artificielle-bnp-paribas-personal-finance); [BNP Paribas](https://group.bnpparibas/en/press-release/bnp-paribas-and-mistral-ai-extend-their-partnership-to-support-the-next-phase-of-generative-ai-deployment-within-the-group) |
| 18 | [Banco Santander](https://www.bankingnewsai.com/banks/santander) | Card-fraud claims automation (Brazil) (in production) · Openbank anti-money-laundering alert models (in production) · Synthetic fraud-network generator (gen-fraud-graph) (announced) | Credit-card eligibility at onboarding (Spain) (in production) · Discrimination testing with counterfactual comparators (mutatis-mutandis) (announced) | Card-fraud claims automation in Brazil: about 95% faster, up to 90% automated, error rate below 1% (Jun 22, 2026) | 3 in production of 5 · score 10 | In production | Jun 25, 2026 | [Banco Santander](https://santander.com/en/stories/santander-turns-its-ai-first-strategy-into-measurable-impact-and-extends-ai-access-to-all-185000-employees); [Banco Santander](https://www.santander.com/en/stories/santander-publishes-ai-projects-under-an-open-source-licence-to-ramp-up-shared-innovation) |
| 19 | [BMO Financial Group](https://www.bankingnewsai.com/banks/bmo-group) | AI alert triage and adverse-media search (in production) · FIS Financial Crimes AI Agent (Claude) (announced) | SmartDecision life-insurance underwriting (in production) · Aura commercial credit agent (rolling out) | Commercial underwriting time target: Roughly 50% lower by fiscal 2028, with 70% of commercial processes powered by AI (Aug 6, 2026) | 2 in production of 4 · score 10 | In production | Sep 1, 2026 | [Financial Post via Yahoo Finance Canada](https://finance.yahoo.com/news/bmo-aims-increase-business-u-200703810.html); [FIS (Business Wire via FT Markets)](https://markets.ft.com/data/announce/detail?dockey=600-202605041800BIZWIRE_USPRX____20260504_BW126906-1); [BMO Insurance (PR Newswire via Financial Post)](https://financialpost.com/press-releases/pr-newswire/bmo-insurance-launches-ai-enhanced-smartdecision-tool-delivering-a-faster-and-more-seamless-path-to-coverage-for-clients-and-advisors) |
| 20 | [Banc of California](https://www.bankingnewsai.com/banks/banc-of-california) | Generative-AI fraud guidance (in production) · Generative-AI voice-authentication fraud exposure (in production) · BSA review support (rolling out) | Venture banking for AI and technology start-ups (in production) · AI-enabled credit analysis and portfolio monitoring (in production) | results not disclosed | 4 in production of 5 · score 9 | In production | Jul 4, 2026 | [Banc of California](https://bancofcal.com/business-insights/the-growing-threat-of-ai-fraud/); [SEC EDGAR](https://www.sec.gov/Archives/edgar/data/1169770/000162828025009438/banc-20241231.htm); [SEC EDGAR](https://www.sec.gov/Archives/edgar/data/1169770/000162828026012946/banc-20251231.htm) |
| 21 | [Ameris](https://www.bankingnewsai.com/banks/ameris) | Customer education on AI-generated fraud (in production) · Photo ID scanning to flag fraudulent IDs (in production) · Voice ID upgrade with automated risk checks (rolling out) | Robotic process automation in insurance premium finance (in production) · AI exposure under fair-lending and equal-opportunity law (in production) | results not disclosed | 4 in production of 5 · score 9 | In production | Sep 24, 2026 | [Ameris Bank](https://www.amerisbank.com/personal/financial-education/ameris-advice/managing-your-money/cybersecurity-and-fraud-trends-for-2025); [Ameris Bancorp (SEC filings)](https://www.sec.gov/Archives/edgar/data/351569/000114036126013511/ny20065565x3_ars.pdf); [Ameris Bank](https://www.amerisbank.com/small-business/loans/insurance-premium-financing) |
| 22 | [Capital One](https://www.bankingnewsai.com/banks/capital-one) | AML and cybersecurity models (in production) · MACAW fraud-call summaries (in production) · ProtectID dealer fraud screen (in production) | no credit use case described | results not disclosed | 4 in production of 4 · score 8 | In production | Sep 24, 2026 | [Capital One](https://www.capitalone.com/tech/ai/transforming-financial-services); [VentureBeat (sponsored by Capital One)](https://venturebeat.com/orchestration/why-capital-one-built-its-multi-agent-ai-platform-around-open-weight-models); [FinTech Breakthrough (GlobeNewswire)](https://www.globenewswire.com/news-release/2026/03/19/3259298/0/en/Capital-One-Auto-s-ProtectID-Wins-2026-Fraud-Prevention-Innovation-Award-for-its-Fraud-Protection-Tool.html) |
| 23 | [Regions](https://www.bankingnewsai.com/banks/regions) | Biometrics and caller monitoring (in production) · Deepfake and AI-scam education (in production) · Adaptive authentication and identity intelligence (in production) | Stress analysis of AI-ecosystem credit exposure (in production) | results not disclosed | 4 in production of 4 · score 8 | In production | Aug 7, 2026 | [Regions Financial (SEC filing)](https://www.sec.gov/Archives/edgar/data/1281761/000128176126000032/rf-2026331xexhibit993fin.htm); [Regions Bank](https://regions.com/insights/wealth/article/deepfake-scams); [Regions Financial (SEC filing)](https://www.sec.gov/Archives/edgar/data/1281761/000128176126000057/a3q26investormeetingmate.htm) |
| 24 | [CIBC](https://www.bankingnewsai.com/banks/cibc-us) | AI in fraud detection and credit monitoring (in production) · Enhanced AML capabilities in the US bank (in production) | DocuMind document intelligence for lending (in production) · Faster, more consistent credit decisions (in production) | results not disclosed | 4 in production of 4 · score 8 | In production | Aug 27, 2026 | [CIBC](https://cibc.com/content/dam/cibc-public-assets/about-cibc/investor-relations/pdfs/annual_meetings/2026/hkc-agm-2026-en.pdf); [CIBC](https://www.cibc.com/content/dam/cibc-public-assets/about-cibc/investor-relations/pdfs/quarterly-results/2025/ar-25-en.pdf); [CIBC via CNW (Yahoo Finance)](https://finance.yahoo.com/technology/ai/articles/cibc-launches-ai-powered-platform-100000321.html) |
| 25 | [City National Florida](https://www.bankingnewsai.com/banks/city-national-florida) | ACH and check Positive Pay, debit block and filtering (in production) · Card Control in the mobile app (in production) · Client education on AI voice-cloning and impersonation scams (in production) | Polly product and pricing engine for residential lending (in production) | results not disclosed | 4 in production of 4 · score 8 | In production | Oct 14, 2026 | [City National Bank of Florida](https://www.citynational.com/treasury-management/fraud-prevention); [City National Bank of Florida](https://www.citynational.com/post/card-control-enhancing-your-digital-wallet-experience); [City National Bank of Florida](https://www.citynational.com/post/what-are-ai-voice-cloning-scams) |
| 26 | [First Horizon](https://www.bankingnewsai.com/banks/first-horizon) | AI-built detection rules for treasury-management payments (in production) | AI file-structure organisation for loan-application documents (in production) · Financial-statement spreading and credit-memo drafting (rolling out) · Automated CRE loan-origination documentation platform (rolling out) | results not disclosed | 2 in production of 6 · score 8 | In production | Oct 1, 2026 | [Investing.com](https://investing.com/news/transcripts/first-horizon-at-raymond-james-conference-strategic-growth-and-challenges-93CH-4538376); [Investing.com](https://investing.com/news/transcripts/first-horizon-at-bank-of-america-conference-strategic-growth-focus-93CH-4500462); [Yahoo Finance](https://finance.yahoo.com/markets/stocks/articles/first-horizon-fhn-q2-2026-111518493.html) |
| 27 | [BankUnited](https://www.bankingnewsai.com/banks/bankunited) | AI-fraud awareness programme (in production) · Commercial fraud-protection products (in production) | no credit use case described | Organisations reporting payment-fraud attempts in 2024: 79% (Jan 9, 2026) | 2 in production of 2 · score 8 | In production | Jun 4, 2026 | [BankUnited](https://bankunited.com/all-resources/why-digital-payments-need-ai-powered-fraud-prevention); [BankUnited](https://bankunited.com/resource-corner/navigating-the-rise-of-ai-generated-phishing-attacks); [BankUnited](https://bankunited.com/resource-corner/video--protect-yourself-from-check-fraud) |
| 28 | [ING](https://www.bankingnewsai.com/banks/ing) | AI-driven KYC and transaction monitoring (in production) · KYC back-office streamlining (rolling out) · Agentic payments with Worldline and Mastercard (pilot) | Instant credit decisions and AI-assisted mortgages (in production) | Instant credit decisions for consumer loans (Netherlands): 65% (Mar 18, 2026) | 2 in production of 4 · score 8 | In production | Oct 5, 2026 | [ING Groep N.V. (SEC filing)](https://www.sec.gov/Archives/edgar/data/0001039765/000162828026011979/ing-20251231.htm); [Computer Weekly](https://www.computerweekly.com/news/366626743/ING-Bank-transforming-operations-through-agentic-AI); [Investing.com](https://ca.investing.com/news/stock-market-news/ing-groep-at-barclays-conference-higher-returns-steady-growth-93CH-4840251) |
| 29 | [NatWest](https://www.bankingnewsai.com/banks/natwest) | Fraud Triage Agent in Cora (in production) · AI-based identity verification for business customers (in production) · Multimodal AI against deepfakes (announced) | no credit use case described | Fraud Triage Agent conversations since launch: Nearly 19,000 (Sep 25, 2026) | 2 in production of 4 · score 8 | In production | Sep 25, 2026 | [NatWest Group](https://natwestgroup.com/news-and-insights/news-room/press-releases/2026/natwest-group-first-to-unveil-next-generation-of-ai-tools-to-hel.html); [NatWest Group plc](https://investors.natwestgroup.com/~/media/Files/R/RBS-IR-V2/results-center/13022026/nwg-annual-report-and-accounts-2025.pdf); [NatWest Group](https://natwestgroup.com/news-and-insights/news-room/press-releases/ai-and-data/2025/jun/natwest-appoints-dr-maja-pantic-as-the-banks-first-chief-ai-rese.html) |
| 30 | [Santander US](https://www.bankingnewsai.com/banks/santander-us) | Deepfake-scam education and AI fraud models (in production) · Openbank AI models for AML alert investigation (group) (in production) · Synthetic fraud-network data (Santander AI Lab) (announced) | Machine-learning auto credit risk on FICO Platform (in production) | results not disclosed | 3 in production of 4 · score 7 | In production | Jun 25, 2026 | [American Banker](https://americanbanker.com/news/santander-creates-fake-ai-ads-for-shopping-scam-awareness); [Banco Santander](https://www.santander.com/en/stories/santander-turns-its-ai-first-strategy-into-measurable-impact-and-extends-ai-access-to-all-185000-employees); [Banco Santander](https://santander.com/en/stories/santander-publishes-ai-projects-under-an-open-source-licence-to-ramp-up-shared-innovation) |

## Which banks lead on fraud AI specifically, and which on credit AI?

The two domains ranked separately by use cases in production, then described; ten each.

| Domain | # | Bank | Use cases in production / described |
| --- | --- | --- | --- |
| Fraud and AML | 1 | [Citi](https://www.bankingnewsai.com/banks/citigroup) | 6 / 6: Faster account opening and onboarding · Behavioural biometrics (BioCatch) |
| Fraud and AML | 2 | [FNBO](https://www.bankingnewsai.com/banks/fnbo) | 5 / 5: Agentic alert triage and enhanced due diligence · Sanctions-screening agent |
| Fraud and AML | 3 | [Commonwealth Bank](https://www.bankingnewsai.com/banks/commonwealth-bank) | 5 / 5: Agentic fraud-pattern detection and rule generation · Scam and fraud detection across payments |
| Fraud and AML | 4 | [Capital One](https://www.bankingnewsai.com/banks/capital-one) | 4 / 4: AML and cybersecurity models · MACAW fraud-call summaries |
| Fraud and AML | 5 | [Zions](https://www.bankingnewsai.com/banks/zions) | 4 / 4: AI in targeted fraud-detection workflows · AI for client authentication |
| Fraud and AML | 6 | [SoFi](https://www.bankingnewsai.com/banks/sofi) | 4 / 4: AI fraud investigations (Poetic) · AI/ML real-time payment-risk platform for clients |
| Fraud and AML | 7 | [TD](https://www.bankingnewsai.com/banks/td-bank-us) | 3 / 4: Fraud defence and claims · AI models in US AML monitoring and investigations |
| Fraud and AML | 8 | [BofA](https://www.bankingnewsai.com/banks/bank-of-america) | 3 / 3: AI fraud-detection models · Behavioural biometrics (BioCatch) |
| Fraud and AML | 9 | [Regions](https://www.bankingnewsai.com/banks/regions) | 3 / 3: Biometrics and caller monitoring · Deepfake and AI-scam education |
| Fraud and AML | 10 | [Western Alliance](https://www.bankingnewsai.com/banks/western-alliance) | 3 / 3: Triangle of Fraud Protection (Digital Disbursements) · AI triage in financial-crimes compliance |
| Credit decisioning | 1 | [Western Alliance](https://www.bankingnewsai.com/banks/western-alliance) | 3 / 3: Credit agreement review and lending lifecycle redesign · AI as an underwriting theme in innovation banking |
| Credit decisioning | 2 | [Stifel Bank](https://www.bankingnewsai.com/banks/stifel-bank) | 3 / 3: Fund banking and wealth-linked lending · Venture lending to AI companies |
| Credit decisioning | 3 | [Zions](https://www.bankingnewsai.com/banks/zions) | 2 / 5: Unstructured-document processing for onboarding and origination · AI in credit review, appraisal review and financial-statement spreading |
| Credit decisioning | 4 | [BMO](https://www.bankingnewsai.com/banks/bmo-us) | 2 / 3: SmartDecision insurance underwriting · ARM automated review triage and portfolio monitoring |
| Credit decisioning | 5 | [SoFi](https://www.bankingnewsai.com/banks/sofi) | 2 / 3: Personal Loan Doc Coach · Oscilar AI risk decisioning for credit strategy |
| Credit decisioning | 6 | [RBC](https://www.bankingnewsai.com/banks/rbc) | 2 / 3: ATOM-based retail credit adjudication · Newcomer credit limit increases |
| Credit decisioning | 7 | [Synchrony](https://www.bankingnewsai.com/banks/synchrony) | 2 / 2: PRISM credit decisioning with cashflow data · PRISM across collections, identity and account management |
| Credit decisioning | 8 | [Popular](https://www.bankingnewsai.com/banks/popular) | 2 / 2: Automatic eligibility decisions on digital loan and card applications (not described as AI) · Fully digital consumer credit origination (not described as AI) |
| Credit decisioning | 9 | [Wintrust](https://www.bankingnewsai.com/banks/wintrust) | 2 / 2: AI credit adjudication at Wintrust Specialty Finance · Automated credit assessment in premium finance |
| Credit decisioning | 10 | [CIBC](https://www.bankingnewsai.com/banks/cibc-us) | 2 / 2: DocuMind document intelligence for lending · Faster, more consistent credit decisions |

## How is this ranking computed?

Score = 2 × fraud, AML/KYC, credit-underwriting and fair-lending use cases in production + 1 × those at other stages + 2 × disclosed result numbers about fraud, AML, approvals, losses or alerts (capped at two). Use cases are the ones tagged with those categories on the bank's page; the tag is applied when the page is written and validated against the use-case list. Ties: systems in production, then assets.

The second table ranks the two domains separately by use cases in production, then by use cases described, then assets.

## What are the caveats?

Fraud AI is close to universal in large banks and often undisclosed in detail: a bank that names only its vendor's fraud model scores one use case, a bank that describes four systems scores four. Credit decisioning is under fair-lending and adverse-action rules, so banks describe it cautiously; several run AI in underwriting without saying so.

Result figures are on different bases (losses avoided, approval-rate lift, alerts reduced, payments screened) and cannot be compared across banks.

## How does this differ from the AI index?

The AI index treats every use case alike; this page keeps only fraud, AML and credit use cases, weights production, and adds the disclosed result where there is one. The AI index and its method: https://www.bankingnewsai.com/banks/ai-index.

## FAQ

### Which bank leads in AI fraud detection?

Citigroup, with 6 fraud and AML AI use cases in production (Faster account opening and onboarding, Behavioural biometrics (BioCatch)).

### Which bank leads in AI credit decisioning?

Western Alliance Bancorporation, with 3 credit AI use cases in production (Credit agreement review and lending lifecycle redesign, AI as an underwriting theme in innovation banking).

### How many banks use AI for fraud detection?

104 of the 140 profiled banks describe a fraud or AML AI system in production; 121 describe one at any stage.

### How many banks use AI to decide credit?

57 describe AI in credit decisioning in production; 87 at any stage. The rest either use conventional scorecards or do not disclose.

### Which banks disclose results from fraud or credit AI?

Zions (Fraud-related loan loss, Q3 2025: $50 million (Oct 20, 2025)), Citi (Cards approval-rate gain from new AI models: +100 basis…), Western Alliance (Fraudulent class-action claims prevented in 2024: 723…), FNBO (Fraud-case investigation with Pindrop Fraud Assist…) and SoFi (Quality reached by Poetic's AI on end-to-end fraud…).

### What rules govern AI credit decisions?

ECOA and Regulation B adverse-action notices, fair-lending law, SR 11-7 model risk, and the EU AI Act's high-risk classification for creditworthiness; the AI fraud detection hub and the regulation tracker cover the documents.

## The other rankings

- [AI leaders](https://www.bankingnewsai.com/banks/leaders/ai): BofA, TD and Citi, by disclosure score, production use cases and named platforms.
- [Technology leaders](https://www.bankingnewsai.com/banks/leaders/technology): JPMorgan, BofA and Citi, by disclosed technology spending, technologists and cloud scale.
- [Proprietary platforms](https://www.bankingnewsai.com/banks/leaders/proprietary-ai-platforms): 99 of 140 banks name one; BofA (Erica (and EricaAssist)), TD (Layer 6 and TD AI Prism), Citi (Citi Stylus Workspaces) lead.
- [Modern technology stack](https://www.bankingnewsai.com/banks/leaders/modern-technology-stack): Arvest, Texas Capital and Deutsche Bank, by disclosed cloud migration, core modernisation and API numbers.
- [Responsible AI](https://www.bankingnewsai.com/banks/leaders/responsible-ai): Citizens Business Bank, Synchrony and Banner Bank, by named AI councils, risk frameworks and responsible-AI officers.
- [AI adoption in 2026](https://www.bankingnewsai.com/banks/leaders/ai-adoption-2026): Amex, SoFi and HSBC, by dated 2026 production rollouts and moves.
- [Research, patents and partnerships](https://www.bankingnewsai.com/banks/leaders/ai-research-patents-partnerships): USAA, TD Bank Group and BofA, by disclosed patents, research deals and model-provider partnerships.
- [Agentic AI](https://www.bankingnewsai.com/banks/leaders/agentic-ai): Amex, Commonwealth Bank and Deutsche Bank (U.S.); 76 of 140 banks have an agent in production.
- [Generative AI with customers](https://www.bankingnewsai.com/banks/leaders/generative-ai-customer-facing): 32 of 140 banks have one in production; BBVA, BNP Paribas and Société Générale lead on stage and scale.
- [All ten rankings](https://www.bankingnewsai.com/banks/leaders)

---

Canonical page: https://www.bankingnewsai.com/banks/leaders/fraud-and-credit-ai
Part of [BankingNewsAI](https://www.bankingnewsai.com/) — a free daily brief on AI in banking, an AI regulation tracker (41 authorities, 238 documents) and AI-strategy profiles of the 120 largest US banks. Markdown versions of every reference page: append `.md` to the page URL; index at https://www.bankingnewsai.com/llms.txt.