Build the platform, rent the models, deploy to everyone — then redeploy the people.
How is JPMorgan using AI?
JPMorgan Chase runs the most publicly documented AI program in banking: a technology budget of nearly $20 billion a year, more than 65,000 technology employees, and a proprietary generative-AI platform, LLM Suite, that has been in the hands of most of its workforce since 2024. Its stated end state, per chief analytics officer Derek Waldron, is a bank in which every employee has a personal AI assistant, every process is run by agents and every client has an AI concierge. In 2026 the program moved from tools to workforce: CEO Jamie Dimon told the July earnings call that AI had cut headcount by 30 to 40 percent in some units, while the bank said it expects to deploy agents that run autonomously for hours before year-end. On the same call Dimon put the count at almost 1,000 AI use cases, about 50 of them 'really important', and the 2025 10-K now lists generative and agentic AI systems, AI-enabled cyber threats and workforce displacement among its risk factors. Leadership changed in July 2026: chief data and analytics officer Teresa Heitsenrether is retiring at year-end, and CTO Scot Baldry takes on her title and the firm's overall AI program.
Lead bank
JPMorgan Chase Bank, N.A.
Assets (lead bank, Jun 30, 2026)
$4.09 trillion
Charter
National bank (OCC)
Headquarters
New York, NY
Flagship platform
LLM Suite
AI leadership
Lori Beer · Teresa Heitsenrether · Derek Waldron · Guy Halamish · Jamie Dimon · Scot Baldry · Manoj Sindhwani · Mark Birkhead · William Patrick Opet · Sumitra Ganesh · Jeremy Barnum · Marianne Lake · Jennifer Piepszak · Mary Callahan Erdoes · Sri Shivananda · Roman Eisenberg · Arvind Joshi · Mike Urciuoli · Umar Farooq
- Technology budget of about $18 billion in 2025 and nearly $20 billion in 2026; more than 65,000 technology employees under global CIO Lori Beer (American Banker, October 2026).
- LLM Suite launched in 2024 as an opt-in corporate ChatGPT and became the firm's flagship gen-AI platform; American Banker's Innovation of the Year 2025.
- Evident's AI index has ranked JPMorgan first among the world's largest banks every year the index has run.
- Agent strategy is explicit and published: a March 2026 engineering post sets safeguards by an agent's capability and 'blast radius'; a June 2026 interview promised long-running autonomous agents in 2026.
- The workforce effect is now disclosed: 30–40% headcount reductions in some units, a 'huge redeployment' plan, and a warning from Dimon that 'you don't uniquely benefit from AI' because rivals have the same tools.
- A named launch partner for Anthropic's restricted Mythos model under Project Glasswing; Dimon calls frontier-model cyber risk 'very heightened'.
- Almost 1,000 AI use cases, 50 'really important' ones across risk, fraud, marketing, hedging, prospecting, note-taking, idea generation and document reading (Q2 2026 call); AI solutions in production doubled in 2025 (February 2026 company update).
- AI leadership reset, July 2026: Heitsenrether retires at end-2026; CTO Scot Baldry becomes chief data and analytics officer and runs the AI program; Manoj Sindhwani takes the data and AI product team; Sumitra Ganesh has led AI research since February.
- LLM Suite went from release in summer 2024 to 200,000 onboarded users in eight months; in February 2026 Dimon put weekly users at 150,000, who estimate about four hours saved a week.
- Use-case counts: about 450 potential cases (March 2025), more than 175 in production in the commercial and investment bank (May 2025), almost 1,000 firmwide with 50 to 60 called significant (May and July 2026).
- The $2 billion AI value figure is stale: Dimon equated annual AI benefit with spend in October 2025, but on the October 2026 call Barnum put annual AI spend nearer $700 million and declined to re-quantify benefits.
- Policy on outside agents: the bank blocks third-party AI agents that log in with customer credentials; Dimon (October 13, 2026) calls it a 'liability transfer' for customers who allow it, while the bank builds its own tools such as Smart Cash, now in pilot.
- The 2025 10-K names disintermediation of customer relationships by AI agents, over-reliance on AI and AI-enabled cyber attacks as risk factors, alongside the workforce displacement language.
- Engineering: a bank-built coding assistant lifted engineer output 10% to 20% (Beer, March 2025); Chase card refactors mainframe code with generative AI inside a sandbox; the 2026 Innovation Week framing is agentic AI across the whole product development lifecycle, with model choice matched to task and cost.
What is LLM Suite?
A proprietary abstraction layer through which third-party models (OpenAI, Anthropic and others) are swapped in and out; rolled out firm-wide in 2024, reported at roughly 250,000 employees with access by late 2025 (American Banker), with the bank itself citing 200,000+ users at the May 2025 investor day and 150,000 weekly users in February 2026.[18][19]
What AI partnerships and patents does JPMorgan Chase have?
JPMorganChase's named AI partners are Anthropic (a Project Glasswing launch organisation since April 2026), OpenAI (GPT-4 behind IndexGPT), Nvidia (an open-source agent safety effort announced in September 2026), OQC and AMD (a London quantum-AI research platform) and Ripcord (lockbox robots). Its entity JPMorgan Chase Bank NA holds a growing set of large-language-model patent filings: two grants (October 2025 and April 2026) and applications published in 2026 on agent frameworks, stock-rating models, hallucination detection and financial prediction. LLM Suite itself is model-agnostic, so no single lab is the exclusive supplier.[48][61][36][45][52][63][64][65][66]
| Item | What it is | Status | Date |
|---|---|---|---|
| Anthropic Project Glasswing | JPMorganChase is one of 12 launch organisations using Claude Mythos Preview for defensive security | Pilot | 2026-04-07[48][5] |
| OpenAI GPT-4 (IndexGPT) | Language model used to build thematic investment baskets | Launched | 2024-05-03[61][62] |
| Nvidia open-source agent safety tools | Shared safety tooling for AI agents; no investment or timetable announced | Announced | 2026-09-28[36] |
| OQC and AMD Quantum-AI Data Centre | Research collaboration in London; JPMorganChase is the first dedicated user | Announced | 2026-06-03[45] |
| Ripcord | AI robotics firm that built the lockbox mail-handling robots | In production | 2026-06-05[52] |
| US 12,443,595 B1 | Verifying SQL queries with large language models (inventors include Sumitra Ganesh) | Granted | 2025-10-14[63] |
| US 12,596,838 B2 | Table question-answering with an LLM while keeping the underlying data private (inventors include Manuela Veloso) | Granted | 2026-04-07[64] |
| US 2026/0050765 A1 | AI agent framework chaining LLMs and a tool registry for workflow analytics | Published application | 2026-02-19[65] |
| US 2026/0111964 A1 | Training and fine-tuning LLMs to predict stock ratings | Published application | 2026-04-23[66] |
| US 2026/0065029 A1 | Perturbation-based zero-shot hallucination reasoning for LLM text | Published application | 2026-03-05[67] |
| US 2026/0010713 A1 | Generative ML framework producing predictive results on financial transactions | Published application | 2026-01-08[68] |
What has JPMorgan done on AI, and when?
- Oct 13, 2026
Q3 2026 call: outside AI agents stay blocked; AI spend put near $700 million[56][55][57]
Dimon says the bank will not let outside agents operate inside accounts and warns of a 'liability transfer' to customers who use them; the Smart Cash pilot is under way; Barnum puts annual AI spend nearer $700 million than the $2 billion cited a year earlier, calls that figure stale and says AI benefits are not yet visible in results.
- Oct 6, 2026
Evident AI Index: first for the fourth consecutive year[58]
Evident's co-CEO attributes the lead to enterprise-wide strategy reaching all 22 lines of business and 10 functions, and says leading banks are building control 'harnesses' for agents.
- Sep 29, 2026
Tops Fortune's expanded AIQ 75 ranking[37]
Fortune and ServiceNow's 2026 AIQ ranking of Fortune 500 companies implementing AI expands from 50 to 75 firms; JPMorgan Chase ranks first overall, ahead of Alphabet, Coca-Cola, Amazon and Nvidia, leading the U.S. commercial-bank cohort (BNY 15, Bank of America 19, Citigroup 51, Wells Fargo 53, Capital One 62).
- Sep 28, 2026
Joins Nvidia-led push for open-source AI agent safety tools[36]
JPMorgan signs on to a shared, open-source safety effort for AI agents led by Nvidia, as the industry works out how to give agents more autonomy without losing control.
- Aug 5, 2026
Kinexys at ~$7 billion a day; federated-learning fraud test[71]
J.P. Morgan reports about $7 billion a day and over $4 trillion since inception on Kinexys; in the prior 12 months JPM Coin launched on Base and Kinexys, BNY, RBC, DeepTempo and NVIDIA completed a federated-learning initiative to improve fraud detection without moving sensitive data.
- Jul 27, 2026
AI leadership reset[34]
Baldry helms the AI program as chief data and analytics officer; Manoj Sindhwani adds the data and AI product team; Sumitra Ganesh has led AI research since Manuela Veloso left in February; the CDAO office shifts from AI infrastructure to business initiatives.
- Jul 22, 2026
JPMorganChase appears in 28% of AI-model answers on consumer banking[59]
A 5W study of 31,500 queries across ChatGPT, Claude, Gemini, Perplexity and Google AI Overviews (January to May 2026) finds the bank cited in 28% of answers against about 17% of US consumer deposits.
- Jul 14, 2026
Q2 2026 call: token spending trivial so far, set to rise[43]
CFO Jeremy Barnum says token expense was a trivial number in the first half, with meaningful acceleration forecast for the second half but a trivial full-year effect; Dimon says the bank does not uniquely benefit from AI because customers will be the beneficiary.
- Jul 14, 2026
- Jul 14, 2026
AI has cut 30–40% of jobs in some units[6]
On the record-quarter earnings call Dimon discloses the reductions, says most staff were offered other roles, and cautions that margins will not uniquely benefit.
- Jul 2, 2026
AI chief Heitsenrether to retire[33]
The chief data and analytics officer, who led the LLM Suite build-out, leaves at year-end; her title and portfolio go to CTO Scot Baldry.
- Jun 18, 2026
Anthropic access cut for Hong Kong staff[13]
Claude models removed from the approved list in Hong Kong, following a similar move by Goldman Sachs.
- Jun 9, 2026
Long-running agents promised for 2026[2]
Waldron: agents will run for an hour or two, act as 'team managers' delegating to sub-agents; AI already lifted private-banking gross sales 20%.
- Jun 5, 2026
Lockbox robots and an LLM assistant[52]
J.P. Morgan Payments says mail-handling robots built with Ripcord handle about 4,000 envelope and document types in a process that covered some 480 million checks and documents a year, now largely automated at over 99.9% accuracy, and staff use an LLM assistant for real-time status.
- Jun 3, 2026
Quantum-AI research collaboration with OQC and AMD[45]
JPMorganChase becomes the first dedicated user of a Quantum-AI Data Centre OQC is building in London, with AMD supplying AI and classical compute; research covers portfolio optimisation and quantum machine learning.
- May 27, 2026
Dimon: 50 or 60 of about 1,000 AI uses are significant; Smart Cash in beta[54]
Speaking at a Bernstein conference, Dimon says the significant ones are saving real money, and the bank is beta-testing Smart Cash, which moves a customer's money between checking and higher-earning brokerage products.
- May 19, 2026
CIO: early AI tools lift productivity up to 30%[9]
Beer tells Bloomberg TV agentic AI promises more, while the bank races to contain new risks.
- Apr 14, 2026
Q1 2026 call: Dimon says AI will not lift the efficiency ratio[42]
He calls it 'a bad idea' to expect AI to improve the efficiency ratio because rivals will deploy it too and pass the benefit on, and points to fraud, scam reduction, prospecting and Connected Commerce as the opportunity side.
- Apr 14, 2026
Q1 2026 call: the 'AI cash tool' is an experiment[31]
Barnum says the consumer AI cash tool from Dimon's letter is 'kind of not even live yet' and aimed at a very small subset of clients; Dimon calls cyber the firm's largest risk and says AI 'has made it worse'.
- Apr 14, 2026
- Apr 7, 2026
Named one of 12 Project Glasswing launch organisations[48]
Anthropic's announcement lists JPMorganChase beside AWS, Apple, Broadcom, Cisco, CrowdStrike, Google, the Linux Foundation, Microsoft, NVIDIA and Palo Alto Networks as users of Claude Mythos Preview for defensive security work.
- Apr 6, 2026
Dimon's shareholder letter: AI touches 'virtually every function'[38]
The letter says AI is not a speculative bubble, will definitely eliminate some jobs while enhancing others, and will almost surely make cyber risk worse; it also previews 'Smart Cash', which moves client cash between accounts for yield.
- Apr 2, 2026
Fence guardrail framework[24]
A data-driven framework that tests LLM use cases for hallucination, topic drift and prompt injection using synthetic data.
- Mar 23, 2026
Engineering post on securing agents[23]
The bank publishes its approach: align safeguards to capability and risk; agents combining untrusted inputs, sensitive data and authority to act require continuous enforcement.
- Feb 24, 2026
Company update: 150,000 weekly LLM Suite users, use cases doubled[39][40]
Dimon says 150,000 people use the platform every week and think they save about four hours a week; Barnum says AI use cases in production doubled and about $600 million of identified efficiencies are partly AI-related; Mary Erdoes says one controls-review AI tool spread from 200 to 3,000 staff.
- Feb 24, 2026
- Feb 23, 2026
Company update deck: AI solutions in production doubled in 2025[28]
The 8-K presentation guides 2026 technology expense to about $19.8 billion, up 10%, and says revenue is the largest contributor to AI benefits, with generative AI the fastest-growing segment.
- Feb 13, 2026
2025 10-K adds a generative and agentic AI risk factor[27]
The annual report lists AI system failures, biased or inaccurate output, agentic systems acting on sensitive data, AI-enabled cyber threats, evolving AI law and AI-driven workforce displacement as risks.
- Feb 12, 2026
- Jan 13, 2026
Q4 2025 call: about $105 billion of 2026 expense, AI included[30]
Management guides adjusted expense of roughly $105 billion to fund technology, payments and AI investment; Dimon says the bank is 'not going to try to meet some expense target' at the cost of falling behind.
- Dec 9, 2025
Productivity doubled with AI, says consumer chief[11]
Marianne Lake: consumer-bank productivity at 6% with AI versus 3% before; operations specialists' productivity expected to rise 40–50%.
- Oct 17, 2025
Chase card uses generative AI to refactor mainframe code[50]
Roman Eisenberg, head of technology for Chase card and connected commerce, describes explaining legacy code with generative AI, generating modern-language code in a sandbox and accepting it only if the same tests pass.
- Oct 14, 2025
Q3 2025 call: AI savings are not the expense target[29]
Pressed on AI-driven productivity, CFO Jeremy Barnum says AI efficiencies are expected but the focus stays on traditional expense discipline and process improvement.
- Oct 13, 2025
Tops the Evident AI Index again[22]
The bank says it continues to lead the 50-bank index across talent, innovation, leadership and transparency.
- Sep 30, 2025
The 'fully AI-connected megabank' blueprint[3]
Waldron describes the end state: personal AI assistants for every employee, agent-powered processes, AI concierges for clients.
- Jun 3, 2025
LLM Suite named American Banker Innovation of the Year[44]
The bank's post says the platform went from zero to 200,000 onboarded users in eight months after its summer 2024 release; Derek Waldron calls it the firm's planned 'AI hub for employees'.
- May 19, 2025
Lake: Chase Auto decisions and funds dealer contracts 'increasingly using AI'[41]
At the 2025 investor day, Consumer & Community Banking CEO Marianne Lake said that for franchise dealers 'we're getting faster at decisioning and funding contracts increasingly using AI', and that 13 million unique users engaged with the Chase car shopping platform in 2024; no tool, model or metric was named.
- May 19, 2025
Investor Day: 200,000 LLM Suite users, 175 AI use cases in CIB[41]
The investor day transcript reports over 200,000 employees with LLM Suite access, more than 175 AI use cases in production in the commercial and investment bank, a 40% fall in KYC unit cost since 2022, and a plan to cut consumer operations headcount about 10% over five years while the business grows more than 25%.
- Mar 13, 2025
Beer: bank-built coding assistant lifts engineer output 10-20%[49]
Global CIO Lori Beer tells Reuters tens of thousands of software engineers gain 10% to 20% from the bank's own coding assistant, that the bank has about 450 potential AI use cases and that Dimon expects 1,000 by 2026.
- Nov 6, 2024
Onyx becomes Kinexys[70]
The blockchain unit Onyx is rebranded Kinexys by J.P. Morgan and JPM Coin System becomes Kinexys Digital Payments; the platform has passed $1.5 trillion in notional value since inception and averages more than $2 billion a day, with payment transactions up 10x year on year.
- Aug 9, 2024
LLM Suite reaches 60,000 employees[18]
CNBC reports the opt-in rollout; by mid-2025 American Banker describes it as available to almost all employees.
- May 3, 2024
IndexGPT thematic baskets launch[61][62]
Built with OpenAI's GPT-4: the model generates keywords for a theme and a separate language model scans news to pick companies. Rui Fernandes, head of markets trading structuring, calls it a first step in a gradual, long-term AI plan for the index line. The name had been filed as a trademark in May 2023.
- Nov 20, 2023
LLMs screen payment validations[69]
J.P. Morgan says it has used AI-powered large language models for payment validation screening for more than two years, cutting false positives and account validation rejection rates by 15–20%.
- Jan 1, 2023
Employees get OpenAI models through an early LLM Suite[3]
The groundwork for the firm-wide platform: a corporate ChatGPT for drafting emails and summarising documents.
- Jan 11, 2018
Chase joins AutoFi: dealer financing decisions online 'often within seconds'[73]
Chase becomes the first national bank on AutoFi's dealer digital-retailing platform, delivering financing terms on dealers' websites often within seconds; the release does not describe the credit decisioning as AI.
- Feb 24, 2015
10-K discloses Justice Department talks on auto dealer markups[72]
The firm says it 'is engaged in discussions' with the Justice Department about potential statistical disparities in markups charged by dealers to borrowers of different races and ethnicities on auto loans it purchased.
Where does JPMorgan use AI today?
| System | What it does | Status | Use case |
|---|---|---|---|
| LLM payment validation screening[69] | Large language models screen payment validations, reducing false positives and improving queue management; account validation rejection rates fell 15–20%. | In production | Fraud detection |
| Federated-learning fraud detection (Kinexys)[71] | With BNY, RBC, DeepTempo and NVIDIA, Kinexys demonstrated federated learning that improves fraud detection while each institution keeps its data in its own environment. | Pilot | Fraud detection |
| LLM Suite and agentic workflows[2][20][21] | Firm-wide gen-AI platform; agents in production for software testing, private-banking research screening and payments; long-running agents planned for 2026. | In production | Generative & agentic AI |
| Call-centre and client assistants[14][3] | Generative AI in call centres for agents serving Chase customers; AI concierge for clients is the stated end state. | Rolling out | Customer-facing chatbots |
| Fraud and AML analytics[25] | Machine-learning monitoring across roughly $10 trillion of daily payment flows, per the bank's own operating description. | In production | Fraud detection |
| AI investing agents (research)[10] | Research built multi-agent systems that beat a 60/40 portfolio in two-decade backtests — a research result, not a product. | Pilot | Trading & capital markets |
| Frontier-model vulnerability testing[5] | Testing Mythos to find vulnerabilities in its own systems as a Glasswing partner. | Pilot | Cybersecurity |
| Firm-wide AI use-case portfolio[32][28] | Almost 1,000 use cases, about 50 of them 'really important' across risk, fraud, marketing, hedging, prospecting, note-taking, idea generation and document reading; AI solutions in production doubled in 2025. | In production | AI governance (general) |
| AI in model risk management and the 10-K risk factor[27] | Models 'including those based upon machine learning or artificial intelligence techniques' sit inside estimations and model risk management; the 10-K names generative and agentic AI failures, bias, transparency and evolving AI law as risks. | In production | Model risk management |
| KYC and transaction-screening AI[41] | KYC unit cost down 40% since 2022 through AI and technology; in Payments, AI models cut manual exceptions by more than 50% and their turnaround time by nearly 75% while volumes rose over 50%. | In production | AML / KYC |
| Coding assistants and AI-assisted mainframe refactoring[49][50] | A bank-built coding assistant lifted engineer productivity 10% to 20%; Chase card documents and rewrites mainframe code with generative AI inside a sandbox, accepting output only when the original tests pass. | In production | AI-generated code & coding agents |
| Smart Cash[54][55] | An AI tool that predicts cash flow and moves customer money between checking and higher-yield brokerage products; in beta in May 2026 and a targeted pilot as of October 13, 2026. | Pilot | Generative & agentic AI |
| Blocking outside AI agents that use customer credentials[56][57] | The bank does not let third parties log in with customers' credentials, so outside agents can at most read data through open-banking connectors and cannot move money; no timeline for direct agent access. | In production | Data & privacy |
| Advisor tools: Connect Coach and Smart Monitor[41] | Connect Coach suggests next-best actions to advisors; it began with 4,600 private-banking advisors, was extended to 7,600 Chase Wealth Management staff in May 2025 and was heading for about 17,000 users by end-2025; Smart Monitor summarises filings and market moves for analysts and portfolio managers. | Rolling out | Generative & agentic AI |
| Consumer travel agent[53] | Lake says the bank aims to have a consumer-facing travel agent in pilot before the end of 2026, while expecting customers not to delegate purchasing to agents soon. | Announced | Customer-facing chatbots |
| AI in auto dealer decisioning and contract funding[41] | Marianne Lake's May 2025 statement is the firm's only public AI claim about auto lending: franchise-dealer contracts are decisioned and funded faster, 'increasingly using AI'. JPMorgan Chase has not named the model, the decisions covered or the share of applications decisioned automatically. | Rolling out | Credit scoring & underwriting |
Use-case links open the tracker's rules-by-use-case matrix: which documents govern each system type.
Which AI vendors and fintechs does JPMorgan work with?
| Company | What JPMorgan uses it for | Status | Category |
|---|---|---|---|
| Kasisto | Named as a customer of Kasisto's KAI conversational AI platform in Kasisto's releases and in American Banker's 2026 acquisition report; the deployment is not described in either source. | In production | Conversational AI |
| ChatGPT Enterprise (OpenAI) | OpenAI models such as GPT-4 swapped into LLM Suite, the bank's in-house gen-AI portal used by 200,000 people. | In production | Enterprise AI assistants |
| Claude (Anthropic) | Claude models on the bank's internal list of approved LLMs under a licensing agreement (withdrawn for Hong Kong staff in June 2026); Project Glasswing launch partner testing Mythos. | In production | Enterprise AI assistants |
| Backbase | Kasisto conversational AI (Kasisto acquired by Backbase June 2026); the deployment is not described in the source | In production | Banking platforms |
| Plaid | Renewed data-access agreement with Plaid, including a pricing structure | Agreement | Fintech · Business finance & infrastructure |
Each vendor and fintech page cites the source for the relationship, and vendor pages list products, pricing and other bank customers. Compare AI tools for banks by category, or read the fintech and neobank AI dossiers.
What has JPMorgan disclosed in numbers?
| Metric | Value | As of |
|---|---|---|
| Kinexys (formerly Onyx) average daily transaction volume | > $2 billion (2024 YTD); ≈ $7 billion (Aug 2026)[70][71] | Aug 5, 2026 |
| Kinexys volume since inception | > $1.5 trillion (Nov 2024); > $4 trillion (Aug 2026)[70][71] | Aug 5, 2026 |
| Account validation rejection rates after LLM screening | −15% to −20%[69] | Nov 20, 2023 |
| Technology budget | ≈ $20 billion (2026); $18 billion (2025)[2][26] | Jun 9, 2026 |
| Technology employees | 65,000+[1] | Oct 1, 2026 |
| LLM Suite access (American Banker figure; the bank cites 200,000+ users and 150,000 weekly) | ≈ 250,000 employees, about half daily users[19] | Apr 29, 2026 |
| Productivity lift from early tools | up to 30%[9] | May 19, 2026 |
| Private-banking gross sales lift | +20%[2] | Jun 9, 2026 |
| Headcount cut in some units | 30–40%[6] | Jul 14, 2026 |
| AI use cases | almost 1,000; about 50 'really important'[32][35] | Jul 14, 2026 |
| 2026 technology expense outlook | ≈ $19.8 billion, +10% year on year[28] | Feb 23, 2026 |
| AI solutions in production | doubled in 2025[28] | Dec 31, 2025 |
| LLM Suite onboarding | 0 to 200,000 users in eight months from the summer 2024 release[44] | Jun 3, 2025 |
| LLM Suite weekly users | 150,000, est. four hours saved per week[39] | Feb 24, 2026 |
| Potential AI use cases | about 450 (Dimon expected 1,000 by 2026)[49] | Mar 13, 2025 |
| AI use cases in production, commercial and investment bank | 175+[41] | May 19, 2025 |
| Engineer productivity from the bank's coding assistant | 10% to 20%[49] | Mar 13, 2025 |
| KYC unit cost since 2022 | -40%[41] | May 19, 2025 |
| Annual AI benefit estimate cited in October 2025 | about $2 billion, roughly equal to spend[55][51] | Oct 7, 2025 |
| Annual AI spend per CFO | closer to $700 million ('a little stale')[55] | Oct 13, 2026 |
| Share of AI-model answers citing JPMorganChase | 28% (vs about 17% of US consumer deposits)[59] | Jul 22, 2026 |
| Lockbox volume automated | about 480 million checks and documents a year, over 99.9% accuracy[52] | Jun 5, 2026 |
“We already have huge redeployment plans for our own people. We have displaced people from AI — and we offer them other jobs.”[4]
Jamie Dimon · Chairman and CEO · Feb 24, 2026
“We've entered now the era of long-running autonomous agents.”[2]
Derek Waldron · Chief Analytics Officer · Jun 9, 2026
“Early AI tools are boosting productivity as much as 30%.”[9]
Lori Beer · Global CIO · May 19, 2026
“I think there's almost 1,000 use cases today, though we'd say that the really important ones are 50 across risk, fraud, marketing, hedging, prospecting, note-taking, idea generation, document reading.”[32]
Jamie Dimon · Chairman and CEO · Jul 14, 2026
“The ultimate beneficiary of AI will be our customers.”[32][35]
Jamie Dimon · Chairman and CEO · Jul 14, 2026
“AI will affect virtually every function, application and process in the company.”[38]
Jamie Dimon · Chairman and CEO · Apr 6, 2026
“I think it's a bad idea to think that you're going to deploy AI and improve your efficiency ratio.”[42]
Jamie Dimon · Chairman and CEO · Apr 14, 2026
“The North Star for LLM Suite is to position it as an AI hub for employees.”[44]
Derek Waldron · Chief Analytics Officer · Jun 3, 2025
“You don't want rogue agents, and there could be thousands of them.”[55]
Jamie Dimon · Chairman and CEO · Oct 13, 2026
“It's not secure, and so to the extent that we're able to block those, we are blocking them to protect customers and protect ourselves.”[57]
Jeremy Barnum · Chief Financial Officer · Oct 13, 2026
Who runs AI at JPMorgan?
Lori BeerLinkedIn ↗
Global Chief Information Officer[16]
Teresa HeitsenretherLinkedIn ↗
Chief Data & Analytics Officer (retiring end-2026; title and portfolio passing to Scot Baldry)[17][33]
Derek WaldronLinkedIn ↗
Chief Analytics Officer[2]
Guy Halamish
COO, Commercial & Investment Bank, with the AI mandate[7]
Scot BaldryLinkedIn ↗
Chief Technology Officer, taking on the Chief Data & Analytics Officer role and the firm's AI program[34][33]
Manoj Sindhwani
CIO for AI/ML and data platforms; also leads the data and AI product team[34]
Mark BirkheadLinkedIn ↗
Firmwide Chief Data Officer[34]
William Patrick Opet
Head of information security, technology controls and technology risk, including AI security[34]
Sumitra GaneshLinkedIn ↗
Head of AI Research (since February 2026)[34]
Jeremy Barnum
Chief Financial Officer; answers analysts on AI spend, token costs and productivity[55][43]
Marianne Lake
CEO, Consumer & Community Banking; aims for a consumer travel agent in pilot before end-2026[53][40]
Jennifer Piepszak
Chief Operating Officer (since early 2025); wrote that the firm is 'as much a technology-driven company as we are a bank'[60]
Mary Callahan Erdoes
CEO, Asset & Wealth Management; sponsor of Connect Coach, Smart Monitor and the controls-review AI tool[41][40]
Roman Eisenberg
Head of technology, Chase card and connected commerce; leads generative-AI mainframe modernisation[50]
Arvind Joshi
Manages the technology budget, including AI investment and token costs[34]
Mike Urciuoli
Head of technology, Asset & Wealth Management[41]
Umar Farooq
Co-Head of J.P. Morgan Global Payments; presented AI models cutting manual exceptions by over 50%[41]
What AI roles is JPMorgan hiring for?
Data as of · 12 AI-titled postings · read from JPMorgan's own careers site, re-checked weekly
12 open AI-related postings on JPMorgan's careers site as of Oct 25, 2026. A snapshot of current openings, not a source: postings close. Compare with other banks on which banks are hiring for AI.
Senior Executive Director of Data & AI
Jersey City, NJ, United States · posted
Data Scientist Director, Executive Director
New York, NY, United States · posted
Executive Director Machine Learning Engineer-MLOps
Palo Alto, CA, United States · posted
NJ, United States · posted
Director of Software Engineering - Agentic AI
Jersey City, NJ, United States · posted
Sr Director of Software Engineering- AI Infrastructure Platforms
Seattle, WA, United States · posted
Principal Software Engineer - AI/ML Data Platforms - Executive Director
LONDON, LONDON, United Kingdom · posted
Executive Director - Senior AI Product Manager
Bengaluru, Karnataka, India · posted
Director of Software Engineering and ML - Agentic Commerce - Executive Director
LONDON, United Kingdom · posted
AI and Data Governance Lead — HR/EX - Vice President
LONDON, LONDON, United Kingdom · posted
APAC Data & AI Compliance and Operational Risk Management – Vice President
Singapore, Singapore · posted
Lead Machine Learning Engineer-MLOps Vice President
NY, United States · posted
Which regulators govern JPMorgan's AI?
| Authority | Why it matters here | Documents |
|---|---|---|
| OCC | Lead bank is a national bank; the revised interagency model risk guidance applies, and generative and agentic AI fall outside it into broader governance. | OCC Bulletin 2026-13 OCC Bulletin 2023-17 |
| Federal Reserve | Holding-company supervisor; SR 26-2 and the third-party guidance govern vendor models such as those behind LLM Suite. | SR 26-2 SR 23-4 |
| CFPB | Consumer lending and Chase chatbots answer to ECOA adverse-action and UDAAP expectations. | ECOA / Regulation B adverse action (15 U.S.C. 1691(d); 12 CFR 1002.9) CFPB Chatbots in Consumer Finance (issue spotlight, 2023) |
| EU AI Act | EU operations: credit scoring is high-risk from December 2, 2027; transparency duties since August 2026. | Regulation (EU) 2024/1689 |
| FSB | As a G-SIB, the FSB's proposed sound practices and its frontier-AI warning are written with banks like this in mind. | FSB AI sound practices consultation (June 2026) FSB Chair's letter to G20 (Aug 2026) |
Side by side: SR 11-7 vs SR 26-2 · OCC vs CFPB on AI lending · PRA SS1/23 vs SR 26-2 · MAS AI Risk Management Guidelines vs SR 26-2 · EU AI Act vs Colorado AI Act · all comparisons.
What does the public record suggest about JPMorgan's AI strategy?
Analysis by BankingNewsAI from the sources cited on this page. Observations, not advice.
01
The disclosure gap is now workforce, not technology
The bank has said what AI does to jobs in some units but not which units or how many people. As other banks are pressed for the same numbers, the July 2026 disclosure sets the reference point analysts will use.
02
Agents outrun the rulebook the bank itself operates under
SR 26-2 excludes agentic AI from model risk management. JPMorgan's published agent-security framework is, in effect, the governance path supervisors asked banks to build for themselves; expect it to be cited back to the industry.
03
Model concentration is a live decision
LLM Suite's design lets models be swapped, and the Hong Kong access change shows the switch is used for jurisdictional reasons, not only performance. Watch which models are approved where.
04
The revenue case is arriving before the cost case
Management's own framing moved in 2026 from savings to sales lift (private banking +20%) and coverage expansion — the metric to watch next is client coverage per banker.
Where does the JPMorgan record go deeper?
7 topic pages under this profile, each answering one question from the same sourced record.
- What is JPMorgan Chase's AI strategy and technology vision?
- What AI research, patents and partnerships does JPMorgan Chase have?
- What technology stack does JPMorgan Chase run its AI on?
- How does JPMorgan Chase govern responsible AI?
- How is AI changing JPMorgan Chase's workforce?
- How does JPMorgan Chase use AI in auto finance?
- What proprietary AI has JPMorgan Chase built in-house?
- Has JPMorgan Chase had a publicly reported AI incident?
Which leader rankings does JPMorgan appear in?
14 of the computed leaders rankings include JPMorgan, each ranked from the same sourced bank pages.
How does JPMorgan compare with its peers?
- How does JPMorgan Chase compare with its peers on AI and technology?
- JPMorgan Chase competitors and alternatives, by business line
Its peer group on this site, by business model and size:
- JPMorgan’s Private Bank Using AI to Make Advisers More Efficient
- JPMorgan Chase, Goldman Sachs underscore security in AI plans
- JPMorgan Is Blocking Outside AI Agents and Taking Its Deposit Token to Ethereum. Here’s What Both Moves Protect
- Schwab and JPMorgan Block Unauthorized AI Agents from Moving Money as Private Ordering Reaches Financial Core
- JPMorgan Blocks Outside AI Agents From Customer Accounts
What is JPMorgan's LLM Suite?
LLM Suite is JPMorgan Chase's proprietary generative-AI platform — a portal and abstraction layer through which large language models from OpenAI, Anthropic and others are made available to employees with the bank's controls. It launched in 2024 on an opt-in basis, reached roughly 250,000 employees by late 2025, and won American Banker's Innovation of the Year award for 2025.
How much does JPMorgan spend on AI?
The bank does not break out AI spending in its filings. Its total technology budget was about $18 billion in 2025 and nearly $20 billion in 2026, according to executives quoted by McKinsey and CNBC. On the October 13, 2026 call CFO Jeremy Barnum put annual AI spend closer to $700 million, called the figure stale and said AI is too diffused to ring-fence.
Has AI cut jobs at JPMorgan?
Yes. On the July 14, 2026 earnings call Jamie Dimon said AI had already reduced headcount by 30 to 40 percent in some of the bank's units, adding that most affected employees were offered positions elsewhere in the firm and that the bank still plans to hire.
Who leads AI at JPMorgan Chase?
Global CIO Lori Beer oversees technology. Chief data and analytics officer Teresa Heitsenrether, who built LLM Suite, retires at the end of 2026, and CTO Scot Baldry is taking on her title and the firm's overall AI program and strategy. Manoj Sindhwani runs AI/ML and data platforms and the data and AI product team, Mark Birkhead is firmwide chief data officer, Sumitra Ganesh heads AI research, Derek Waldron is chief analytics officer and Guy Halamish carries the AI mandate as COO of the commercial and investment bank.
How many people use JPMorgan's LLM Suite?
LLM Suite reached 200,000 onboarded users within eight months of its summer 2024 release, according to the bank, and the May 2025 investor day put access above 200,000 employees. At the February 2026 company update Jamie Dimon said 150,000 people use it every week and think they save about four hours a week. The bank does not tie that time saving to headcount. Derek Waldron describes the platform's aim as an AI hub for employees, with internal APIs now letting staff build generative AI into their own workflows.
Does JPMorgan Chase let outside AI agents access customer accounts?
No. On October 13, 2026 a spokeswoman confirmed to CNBC that the bank blocks external AI agents from operating for retail customers and does not allow third parties to log in with customers' banking credentials. Agents can read some data, such as balances, through open-banking providers such as Plaid, but cannot send money or pay bills. Jamie Dimon warned of a 'liability transfer' to customers who use them, and CFO Jeremy Barnum said the bank is working on secure ways to enable agents, with no timeline.
Is JPMorgan's AI spending paying for itself?
The bank no longer gives a clean answer. In October 2025 Jamie Dimon said annual AI benefit of about $2 billion roughly matched what the bank spends on AI. On the October 13, 2026 call CFO Jeremy Barnum put annual AI spend closer to $700 million, called that stale, and said he is reluctant to re-quantify benefits because the old estimate included traditional machine learning. He added that AI gains are not yet visible in quarterly results, partly because token costs offset productivity.
- The Most Powerful Women in Banking, #5, Lori Beer, JPMorganChase
- JPMorgan Chase plans to deploy more powerful AI agents this year
- Here's JPMorgan Chase's blueprint to become the world's first fully AI-powered megabank
- Jamie Dimon says AI is already reshaping JPMorgan Chase's workforce as bank plans 'huge redeployment'
- Jamie Dimon says Anthropic's Mythos reveals 'a lot more vulnerabilities' for cyberattacks
- Bank earnings takeaways: from Goldman Sachs' SpaceX IPO fees to JPMorgan's AI job cuts
- JPMorgan's commercial and investment bank names Halamish as COO to lead AI strategy, memo shows
- JPMorgan Promotes Halamish With Mandate to Accelerate AI Rollout
- JPMorgan's CIO Sees Leadership Test as AI Fuels Major Shift
- JPMorgan Builds AI Agents That Beat 60/40 Portfolio in Backtests
- US bank executives say AI will boost productivity, cut jobs
- Banks in Asia brace for complex cyber threats from frontier AI
- JPMorgan Chase cuts off Anthropic access for its Hong Kong staff
- The Rise of Artificial Intelligence at JPMorgan
- Jamie Dimon Dismisses Fears Over How AI Will Hit JPMorgan
- Lori Beer is turning JPMorganChase into an AI-native bank, and she's got $20 billion to do it
- The Most Powerful Women in Banking, No. 8, Teresa Heitsenrether, JPMorganChase
- How JPMorganChase democratized employee access to gen AI
- Driving AI Use Cases and Adoption (Women in Banking 2025 session)
- Translating technology from cost to competitive edge (Gill Haus)
- JPMorgan Payments' agentic commerce play
- JPMorganChase continues to lead the world's top banks in AI maturity
- Securing the next generation of AI agents
- Strengthening LLM guardrails with synthetic data generation
- Future-Proofing the Company and Our Operations (COO letter, 2024 annual report)
- JPMorgan Chase's Derek Waldron on building an AI-first bank culture
- JPMorgan Chase & Co. Annual Report on Form 10-K for the year ended December 31, 2025
- 2026 Company Update presentation slides (Form 8-K, Exhibit 99.2)
- 5 Revealing Analyst Questions From JPMorgan Chase's Q3 Earnings Call
- JPMorgan Chase & Co. Q4 Earnings Call Highlights
- JPMorgan Chase & Co. Q1 Earnings Call Highlights
- JPMorgan Chase (JPM) Q2 2026 Earnings Call Transcript
- JPMorgan AI chief Teresa Heitsenrether retiring in 2026
- JPMorgan's AI leadership is getting a reset. These are the key players.
- Banks report operational changes driven by AI adoption
- JPMorgan Stock Slips 1% as Bank Agents Get an Open Safety Layer
- How bottom-line-focused financial firms are finding ROI in AI
- Jamie Dimon's Letter to Shareholders, Annual Report 2025
- 2026 Company Update Transcript - Chairman and CEO Q&A
- 2026 Company Update Transcript (full event)
- 2025 Investor Day Transcript - Full Transcript
- 1Q26 Earnings Call Transcript
- 2Q26 Earnings Call Transcript
- LLM Suite named 2025 'Innovation of the Year' by American Banker
- OQC, JPMorganChase and AMD commence research collaboration to develop new quantum-AI platform
- Innovation for business impact: Key takeaways from JPMorganChase's 11th Innovation Week
- AI in Practice: Building Workplace Skills for the Future
- Project Glasswing: Securing critical software for the AI era
- JPMorgan engineers' efficiency jumps as much as 20% from using coding assistant
- Chase enlists generative AI to drive modernization
- Banks chase AI-fueled efficiencies
- JPMorgan Chase taps AI to process checks
- JPMorgan's Lake sees longer road to agentic commerce adoption
- JPMorgan CEO: AI may create things 'we're going to lose at'
- JPMorgan CEO Dimon warns customers of 'liability transfer' with AI
- JPMorgan Chase blocks outside AI agents like Muse, Instinct from directly accessing accounts
- JPMorgan warns of risks to consumers who rely on AI agents
- JPMorganChase still tops AI ranking as big-bank lead widens
- JPMorgan owns AI answers. Here's how rivals can compete.
- The Most Powerful Women in Banking, #4, Jennifer Piepszak, JPMorganChase
- JPMorgan unveils IndexGPT in next Wall Street bid to tap AI boom
- JPMorgan taps ChatGPT for new thematic investment suite
- US12443595B1 - Method and system for verification with large language models for data querying
- US12596838B2 - Method and system for performing table question-answering tasks while preserving data security
- US20260050765A1 - Method and system for an AI agent framework performing workflow analytics
- US20260111964A1 - Training and fine-tuning large language models in generating stock ratings predictions
- US20260065029A1 - Perturbation-based zero-shot hallucination reasoning for LLM-generated text
- US20260010713A1 - Generative machine learning framework generating predictive results
- How AI will make payments more efficient and reduce fraud
- Introducing Kinexys by J.P. Morgan
- How Blockchain Moved From Pilots to Production-Grade Payments
- JPMorgan Chase & Co. Annual Report on Form 10-K for the year ended December 31, 2014
- Chase Partners with AutoFi to Deliver Digital Car-Buying for Dealerships across the Country
When JPMorgan 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