Disclosed numbers · computed from bank profiles; nothing on this page is written by a model. All ranked answers: /banks/leaders
On the public record as of Oct 28, 2026, 38 of the 141 large banks profiled disclose a dated figure for hours saved, time freed or productivity gained from AI. Ranked by matching figures, the top 5 are CIBC (5 figures), NatWest (4 figures), Customers Bank (4 figures), Barclays (4 figures) and BBVA (3 figures). CIBC's strongest evidence: Developers with AI capabilities achieving 20% productivity benefits: 4,000+ (as of Aug 27, 2026).
141 banks profiled (120 US, 21 non-US) · 38 with matching evidence · ranked answer, not a comparison
Which of the large banks disclose hours saved or productivity gains from AI?
| # | Bank | Matching figures | Figure (as reported) | Status | Date | On this site | Sources |
|---|---|---|---|---|---|---|---|
| 1 | CIBC | 5 | Developers with AI capabilities achieving 20% productivity benefits: 4,000+ (as of Aug 27, 2026) | Disclosed | Aug 27, 2026 | CIBC profile → | CIBC ↗ · Yahoo Finance ↗ |
| 2 | NatWest | 4 | Meeting-preparation hours saved in Commercial Mid-Market: 22,000 hours by 1,200 colleagues using AI (as of Jul 27, 2026) | Disclosed | Jul 27, 2026 | NatWest profile → | NatWest Group plc ↗ · NatWest Group plc (company announcement) ↗ |
| 3 | Customers Bank | 4 | Productivity gain from AI-powered KYC screening: 50% (as of Jul 24, 2026) | Disclosed | Jul 24, 2026 | Customers Bank profile → | American Banker ↗ · Customers Bancorp (SEC filings) ↗ |
| 4 | Barclays | 4 | Developers with AI tools: 19,000; early-stage productivity gains of about 15% (as of Feb 10, 2026) | Disclosed | Feb 10, 2026 | Barclays profile → | Yahoo Finance (Quartr) ↗ · Barclays PLC (SEC 20-F) ↗ |
| 5 | BBVA | 3 | Employees using generative AI on a recurring basis / time saved: 75% / about 2.4 hours a week (as of Oct 6, 2026) | Disclosed | Oct 6, 2026 | BBVA profile → | OpenAI ↗ · BBVA ↗ |
| 6 | First Horizon | 2 | Staff hours saved by the AI file-organisation tool: hundreds, within three months of use (as of Oct 1, 2026) | Disclosed | Oct 1, 2026 | First Horizon profile → | American Banker ↗ |
| 7 | Citizens | 2 | Average engineer productivity gain (Reimagine the Bank): 30%-40% more productive than a year earlier (as of Sep 29, 2026) | Disclosed | Sep 29, 2026 | Citizens profile → | American Banker ↗ · Banking Dive ↗ |
| 8 | BMO | 2 | Lumi: new-employee productivity gain: +17% (as of Aug 25, 2026) | Disclosed | Aug 25, 2026 | BMO · ai workforce → | BMO AI Hub ↗ · Motley Fool (Yahoo Finance) ↗ |
| 9 | Schwab | 2 | Developer team productivity gain over the past year: 15–20% (as of Jul 21, 2026) | Disclosed | Jul 21, 2026 | Schwab profile → | The Motley Fool (Yahoo Finance) ↗ · InvestmentNews ↗ |
| 10 | ConnectOne | 2 | Frontline hours targeted for release per banker: About 1,000 of roughly 2,000 working hours a year (as of Jul 14, 2026) | Disclosed | Jul 14, 2026 | ConnectOne profile → | nCino (GlobeNewswire) ↗ · American Banker ↗ |
| 11 | Western Alliance | 2 | Financial-crimes compliance triage: 60–80% faster (internal projection) (as of May 12, 2026) | Disclosed | May 12, 2026 | Western Alliance profile → | Western Alliance Bancorporation (SEC filings) ↗ |
| 12 | Community Bank N.A. | 2 | 2025 incentive-plan technology objective: hours saved: 65,000 hours from information-technology initiatives (as of Mar 27, 2026) | Disclosed | Mar 27, 2026 | Community Bank N.A. profile → | MarketBeat (Yahoo Finance) ↗ · Community Financial System (SEC filings) ↗ |
| 13 | Bread Financial | 2 | Engineering throughput gain from AI: About 30% faster (as of Mar 11, 2026) | Disclosed | Mar 11, 2026 | Bread Financial profile → | Investing.com ↗ |
| 14 | Barclays US | 2 | Query-response speed-up from the US app's conversational AI tool: 95% faster (chief executive's figure at launch) (as of Feb 10, 2026) | Disclosed | Feb 10, 2026 | Barclays US profile → | Yahoo Finance (Quartr) ↗ · SEC (Barclays PLC 20-F) ↗ |
| 15 | OCBC | 2 | Gen AI applications deployed: Over 30, with up to 50% productivity gains for some tasks (as of Mar 25, 2025) | Disclosed | Mar 25, 2025 | OCBC profile → | Oversea-Chinese Banking Corporation ↗ · Oversea-Chinese Banking Corporation ↗ |
| 16 | BofA | 1 | Developers on coding agents: ≈ 20,000; 15% to 20% productivity gain (as of Sep 23, 2026) | Disclosed | Sep 23, 2026 | BofA · ai workforce → | Banking Dive ↗ |
| 17 | Truist | 1 | AI Call Summaries generated / hours saved: 4.5 million+ summaries; 36,000+ teammate hours in Q2 2026 (as of Sep 9, 2026) | Disclosed | Sep 9, 2026 | Truist · ai workforce → | Truist ↗ |
| 18 | M&T | 1 | Time saved per call (gen-AI summaries): ≈ 6 minutes (as of Sep 2, 2026) | Disclosed | Sep 2, 2026 | M&T profile → | Fast Company ↗ |
| 19 | BOK Financial | 1 | Time saved per initial trust review with Vanilla (vendor-reported): ≈ 4 hours on average (as of Aug 13, 2026) | Disclosed | Aug 13, 2026 | BOK Financial profile → | Vanilla (vendor case study) ↗ |
| 20 | SouthState | 1 | Estimated Copilot time saved (company estimate): ≈ 100 FTE-equivalents a month (as of Aug 3, 2026) | Disclosed | Aug 3, 2026 | SouthState · ai workforce → | SouthState Correspondent Division ↗ |
| 21 | BMO Financial Group | 1 | Wealth AI tools: assisted hours a year: More than 400,000 (as of Jul 27, 2026) | Disclosed | Jul 27, 2026 | BMO Financial Group profile → | BMO AI Hub ↗ |
| 22 | Deutsche Bank (U.S.) | 1 | Developers with AI coding assistants / time saved: More than 15,000 / 1.5 to 2.5 hours a week (as of Jul 14, 2026) | Disclosed | Jul 14, 2026 | Deutsche Bank (U.S.) · ai workforce → | Deutsche Bank ↗ · Deutsche Bank ↗ |
| 23 | Deutsche Bank | 1 | Developers with AI coding assistants / time saved: More than 15,000 / 1.5-2.5 hours a week; 10x on specific tasks (as of Jul 14, 2026) | Disclosed | Jul 14, 2026 | Deutsche Bank profile → | Deutsche Bank ↗ · Deutsche Bank ↗ |
| 24 | Morgan Stanley | 1 | Controller hours saved a week by FIXR: ≈ 1,500 (across ≈ 100 controllers) (as of Jun 30, 2026) | Disclosed | Jun 30, 2026 | Morgan Stanley · ai workforce → | VentureBeat ↗ |
| 25 | Banco Santander | 1 | Card-fraud claims automation in Brazil: about 95% faster, up to 90% automated, error rate below 1% (as of Jun 22, 2026) | Disclosed | Jun 22, 2026 | Banco Santander profile → | Banco Santander ↗ |
| 26 | SoFi | 1 | Faster deployment of credit-risk strategies after moving to Oscilar: 50% faster; 30%+ faster decisioning (as of Jun 3, 2026) | Disclosed | Jun 3, 2026 | SoFi profile → | Oscilar (PR Newswire) ↗ |
| 27 | Citi | 1 | Engineering capacity freed by AI code review: about 100,000 hours a week (as of May 7, 2026) | Disclosed | May 7, 2026 | Citi · ai workforce → | Citigroup ↗ |
| 28 | Pinnacle | 1 | ChatPFP usage: 18,000 banker questions answered; over 3,000 hours saved (as of Apr 23, 2026) | Disclosed | Apr 23, 2026 | Pinnacle profile → | The Motley Fool (earnings call transcript) ↗ |
| 29 | BNY | 1 | Settlement inquiries addressed by AI / investigation speed-up: > 10% / > 80% faster (1Q26 vs 1Q25) (as of Apr 16, 2026) | Disclosed | Apr 16, 2026 | BNY · ai workforce → | BNY (SEC filing) ↗ |
| 30 | JPMorgan | 1 | LLM Suite weekly users: 150,000, est. four hours saved per week (as of Feb 24, 2026) | Disclosed | Feb 24, 2026 | JPMorgan · ai workforce → | JPMorganChase ↗ |
| 31 | U.S. Bank | 1 | Code-review time saved by the GenAI review assistant: 75% (as of Feb 23, 2026) | Disclosed | Feb 23, 2026 | U.S. Bank · ai workforce → | U.S. Bank ↗ |
| 32 | Standard Chartered | 1 | Relationship-manager time saved on client-message drafting (bank-reported): Up to 50% (as of Jan 14, 2026) | Disclosed | Jan 14, 2026 | Standard Chartered profile → | Standard Chartered PLC ↗ |
| 33 | UBS (US) | 1 | Advisor time saved on meeting preparation (estimate): About 10,000 hours a month (as of Dec 1, 2025) | Disclosed | Dec 1, 2025 | UBS (US) · ai workforce → | Financial Planning ↗ |
| 34 | Lloyds | 1 | Copilot time saved per colleague: 46 minutes a day (survey of 1,000 users) (as of Oct 21, 2025) | Disclosed | Oct 21, 2025 | Lloyds profile → | Lloyds Banking Group ↗ · The Register ↗ |
| 35 | Zions | 1 | Speed-up for products relying on the new core: 20% faster (as of Sep 24, 2025) | Disclosed | Sep 24, 2025 | Zions profile → | American Banker ↗ |
| 36 | Northern Trust | 1 | Developer time saved with GitHub (internal estimate): About 31% (as of Sep 9, 2025) | Disclosed | Sep 9, 2025 | Northern Trust profile → | Investing.com ↗ |
| 37 | Amex | 1 | Travel counselors using the AI assistant: 5,000 in 19 markets; over 85% report time saved (as of Apr 2, 2025) | Disclosed | Apr 2, 2025 | Amex profile → | VentureBeat ↗ |
| 38 | Fulton | 1 | Agent time saved per call by AI summarisation: 3–5 minutes (as of Nov 20, 2024) | Disclosed | Nov 20, 2024 | Fulton profile → | American Banker ↗ |
Each row quotes one record from bank profiles; the “On this site” link opens the bank's own page or sub-page with every source. Banks absent from the table have nothing matching on the record this site holds.
How is this ranked?
Data: the 141 bank profiles on this site (lib/banks), every entry dated and sourced: "The numbers" table, each figure quoted as reported with its as-of date and source.
Rule: Figures whose label or value matches the page's pattern (see the code). Score = matching figures; ties by the newest as-of date.
Counts: 38 of 141 profiled banks qualify; 49 distinct sources cited in the table. Data checked 2026-10-28; ranking template updated 2026-10-30.
What are the caveats?
Figures are as each bank reported them and are not normalised: a share of employees and a count of prompts are not comparable.
Banks absent from the table have nothing matching on the record this site holds, which is not evidence that they do nothing. Each row links the bank's own page, where every entry carries its sources.
Which bank ranks first?
CIBC Bank USA leads with 5 figures: Developers with AI capabilities achieving 20% productivity benefits: 4,000+ (as of Aug 27, 2026). NatWest is second with 4.
How many large banks disclose AI hours saved or productivity gains?
38 of the 141 profiled banks (120 US, 21 non-US groups). Among those listed, 29 are US banks and 9 non-US groups.
What counts, and what does not?
Figures whose label or value matches the page's pattern (see the code). Score = matching figures; ties by the newest as-of date. Source: the 141 bank profiles on this site (lib/banks), every entry dated and sourced: "The numbers" table, each figure quoted as reported with its as-of date and source.
Is this a measure of capability?
No. Every row is a sourced fact on the public record: what the bank, a regulator, a filing or an independent outlet has put on the record. The page does not rate how well anything works.
When was this checked, and how often does it update?
The data behind the table was last checked on 2026-10-28 (bank profiles). The page recomputes whenever that dataset changes and is republished with the site; it is never edited by hand.
When any of these banks moves on AI, you'll read it the next morning.
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