How does Simmons Bank use AI and scoring to fight fraud, and does it use AI in credit decisions?
Simmons Bank's fraud record rests on two systems built by its digital team under chief digital officer Alex Carriles. The first is an in-house mobile-deposit scoring algorithm, built with Jack Henry, that scores the customer rather than the check: Jack Henry's 2022 case study says the bank scored every customer every day and that mobile-deposit dollars rose 59% from Q3 2020 to Q3 2021, and its 2025 case study reports a 96% reduction in fraud losses, to less than 1 bp, with limits raised from $1,500 per month to up to $150,000. The second is digital account opening on Jack Henry's OpenAnywhere, adopted in early 2021 with identity proofing from IDScan.net, which Jack Henry and Bank Automation News describe as AI-powered: an ID scan, a selfie liveness check and DMV database checks, after which Carriles said in November 2022 that the bank had seen 'fraud attempts pretty much drop down to zero'. In April 2024 he told the Wall Street Journal the bank was changing its protocols because AI can generate fake driver's licenses: customers now photograph the license in the app and take a selfie while the app asks them to look left, right, up or down. Neither case study calls the deposit score machine learning, and Simmons has disclosed no AI model for credit decisions. Its 2025 10-K and 2026 10-Qs treat AI as a risk, including cyber-attacks using generative AI and deepfakes.[1][2][3][4][5][6][7]
Data as of Nov 17, 2026 · Simmons Bank profile updated Nov 18, 2026 · 14 sources cited on this page · 141 banks profiled on this site · Markdown
Which fraud controls has Simmons Bank described?
| Control | What it does | Technology | Date |
|---|---|---|---|
| Mobile-deposit customer scoring | Scores each customer every day for mobile-deposit eligibility and limits; limits are reduced or canceled when a customer stops behaving responsibly and raised for a good relationship | In-house development, Banno API integration and custom code | 4 May 2022[1] |
| Monthly limit recalculation | Each customer's limit is recalculated monthly; the score sets eligibility, deposit limits, risk level and required deposit holds, and a SilverLake customization applies the holds at deposit time | jXchange, JHA SmartPay Remote Deposit Anywhere, Enterprise Payment Solutions | 10 September 2025[2] |
| Score-based fraud review | The fraud department is more selective about which mobile deposits it reviews, choosing them by the customer's score | Same in-house score | 10 September 2025[2] |
| Digital ID verification at account opening | The customer photographs the front and back of an ID and passes a selfie likeness check; back-end queries of DMV and Identifraud databases confirm identity, address and account eligibility | JHA OpenAnywhere with IDScan.net, described as an AI-powered identity verification platform | Adopted early 2021; case study 7 August 2023[3] |
| Driver's license validation | Scans the license or state ID, takes a picture of the user and sends the data to the American Association of Motor Vehicle Administrators; the usual identity questions about loans and mortgages were dropped | IDscan integrated with OpenAnywhere through APIs | 20 July 2021[8] |
| Liveness and anti-spoofing | A selfie liveness check with anti-spoofing measures against techniques such as deepfakes, plus ID validation checks and calls to Department of Motor Vehicles databases | IDScan.net on Jack Henry's OpenAnywhere | 15 November 2022[9] |
| Head-movement selfie check | Customers photograph the license through the app instead of uploading an existing picture, then take a selfie while the app asks them to look left, right, up or down, because AI can generate fake driver's licenses | IDScan.net | 3 April 2024[5] |
| Device and transaction monitoring | Two-factor authentication when someone logs in from an unfamiliar device; all transactions monitored for signs of fraud | Not named | 15 June 2022[10] |
What results has Simmons reported from these controls?
| Measure | Figure | Period | Reported by |
|---|---|---|---|
| Mobile-deposit accounts | Up 32% | Q3 2020 to Q3 2021 | Jack Henry case study, 2022[1] |
| Mobile-deposit dollars | Up 59%; dollar volume tripled in the first year; some customers' limits rose 20 times | Q3 2020 to Q3 2021 | Jack Henry case study, 2022[1] |
| Fraud on the mobile-deposit platform | Down 96% | Reported June 2022 | American Banker[10] |
| Fraud losses | Down 96%, to less than 1 bp, saving hundreds of thousands of dollars | Reported 2025 | Jack Henry case study[2] |
| Mobile-deposit limits | From $1,500 per month to up to $150,000, with funds available in five seconds, 24 hours a day | Reported 2025 | Jack Henry case study[2] |
| Synthetic identity fraud | 'Near zero' after the bank layered on additional third-party checks; onboarding takes less than 5 minutes | Case study dated 2023 | Jack Henry[3] |
| Account-opening fraud attempts | 'Pretty much drop down to zero' after the ID scan integration, per Carriles | Reported November 2022 | Bank Automation News[4] |
| IDs validated on Jack Henry's platform | More than 100,000 customers across Jack Henry institutions, not Simmons alone | Since launch, as of November 2022 | IDScan.net release on Finextra[9] |
Which parts of this are AI, and what does Simmons not disclose?
The AI on the record is IDScan.net's. Jack Henry's 2023 case study calls it an AI-powered identity verification platform supplying ID parsing, face matching and identity proofing, and Bank Automation News wrote in 2022 that its AI-powered technology verifies a license's authenticity through a digital image analysis.[3][4]
The mobile-deposit score is described as an in-house scoring and qualification algorithm and a risk-scoring model; neither Jack Henry case study says it uses machine learning or AI.[2][1]
The one lending system Simmons has described in detail is Finastra's LaserPro loan-document engine, which maps over 85% of the data fields from its loan origination system; Finastra says the product automates the manual financial spreading process, and the case study describes no credit-decision model.[11]
The filings treat AI as a risk, not a tool: the 2025 10-K, filed 25 February 2026, names agentic AI, third-party models and generative models that may produce incorrect outputs, and the second-quarter 2026 10-Q lists cyber-attacks using generative artificial intelligence and deepfakes.[6][7]
What does the Simmons Bank profile record on this topic?
The rows below are the entries of the Simmons Bank AI strategy profile that bear on this page, as of Nov 18, 2026; the profile holds the full record of 27 dated moves, 10 use cases and 33 sources.
| Use case | What the record shows | Status | Source |
|---|---|---|---|
| Generative-AI and deepfake attack risk | 10-Q cyber risk factor names generative AI and deepfakes among attack techniques. | In production | [7] |
| Mobile-deposit customer scoring | In-house scoring and qualification algorithm built with Jack Henry: evaluates every customer daily, recalculates limits monthly, sets deposit eligibility, limits, risk level and holds; fraud losses down 96% to under 1 basis point; the source does not say whether the model is machine learning. | In production | [2][12] |
| CISO-led information security programme with board IT Committee reporting | Security operations, identity and access, engineering and architecture groups under a CISO who reports to the CIO; cyber KRIs go to the enterprise risk management committee and the Simmons Bank board IT Committee. | In production | [6] |
| Board and executive risk oversight of IT, cyber and vendors | Board Risk and Compliance Committee oversight of fraud, third-party vendor, cybersecurity and IT risk, with an Enterprise Risk Management Committee of senior executives and the chief risk officer administering the programme; no AI-specific committee is named. | In production | [13] |
| Date | Move | Detail | Source |
|---|---|---|---|
| Aug 5, 2026 | Q2 10-Q: generative AI and deepfakes in cyber-attacks | Cyber-attack techniques are 'increasingly sophisticated, including through the use of generative artificial intelligence and deepfakes'. | [7] |
| May 6, 2026 | Q1 10-Q repeats the agentic-AI risk factor | The 10-Q carries the agentic-AI and third-party-model risk factor and the cyber-attack language on generative AI and deepfakes. | [14] |
| Apr 8, 2026 | 2026 proxy statement: no AI mention; risk committee covers IT, cyber and vendors | The board's Risk and Compliance Committee covers fraud, third-party vendors, cybersecurity and information technology; the chief risk officer runs the risk programme; executive officers listed include no CIO, CTO or CDO. | [13] |
| Sep 10, 2025 | Jack Henry Cobalt Award to Simmons for mobile-deposit scoring | Simmons won the Changemaker category for a 96% reduction in fraud losses; the in-house customer-scoring algorithm sets mobile-deposit eligibility, limits, risk level and holds. | [12][2] |
Does Simmons Bank use AI to detect fraud?
At account opening, yes. Since early 2021 its digital account opening on Jack Henry's OpenAnywhere has used IDScan.net, which Jack Henry calls an AI-powered identity verification platform, for ID parsing, face matching and liveness checks. Its best-known fraud tool, the mobile-deposit score, is described as an in-house algorithm, not as AI.
How does Simmons Bank's mobile-deposit scoring work?
It scores the customer rather than the check. The in-house algorithm, built with Jack Henry, evaluates every customer daily and recalculates limits monthly; the score sets eligibility, limits, risk level and deposit holds. Jack Henry reported a 96% fall in fraud losses, to less than 1 bp, in 2025.
How does Simmons protect digital account opening against deepfakes?
Customers photograph their driver's license through the Simmons app and then take a selfie. In April 2024 Alex Carriles told the Wall Street Journal the app asks users to look left, right, up or down, since a generic AI deepfake may not do the same; IDScan.net's liveness check includes anti-spoofing measures.
Does Simmons Bank use AI for credit decisions?
Simmons has disclosed no AI or machine-learning credit model. The lending technology it has described is Finastra's LaserPro loan-document engine, which maps over 85% of data fields from its loan origination system, and its 10-K and 10-Qs discuss AI only as a risk.
What fraud results has Simmons Bank reported?
Mobile-deposit fraud down 96% (American Banker, June 2022; Jack Henry, 2025), fraud losses under 1 bp, synthetic identity fraud near zero after added third-party checks (Jack Henry, 2023), and account-opening fraud attempts 'pretty much' down to zero (Bank Automation News, November 2022).
- Simmons Bank is blazing a trail with the Banno Digital Toolkit (case study)
- Simmons Bank changes the game with mobile deposit scoring
- Simmons Bank reduces fraud and improves digital onboarding with identity verification (case study)
- Simmons Bank leverages ID verification solution
- Deepfakes Are Coming for the Financial Sector
- Simmons First National Corporation Annual Report on Form 10-K for the fiscal year ended December 31, 2025
- Simmons First National Corporation Quarterly Report on Form 10-Q for the quarter ended June 30, 2026
- From Simmons First, an account just for young crowd
- IDScan.net to provide identity validation for banks on the Jack Henry platform
- How Alex Carriles attracted Gen Z to Simmons Bank
- LaserPro helps rapidly-growing Simmons Bank centralize its loan applications while preserving its local community feel (case study)
- Jack Henry Announces 2025 Cobalt Award Winners
- Simmons First National Corporation 2026 Proxy Statement (Schedule 14A)
- Simmons First National Corporation Quarterly Report on Form 10-Q for the quarter ended March 31, 2026
When Simmons Bank moves on AI, you'll read it here first.
when one of the banks we cover 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