JPMorgan Chase & Co. is one of the world’s largest financial institutions, reporting $4.425 trillion in total assets, $2.559 trillion in deposits, $57.0 billion in net income, and $182.4 billion in total net revenue in 2025. It serves millions of customers across consumer banking, commercial banking, payments, investment banking, and asset management, and ended the year with $288.5 billion in CET1 capital and about $1.5 trillion in liquidity sources.
The firm has long been a leader in financial technology, investing heavily in AI to drive efficiency, innovation, and risk management. With an annual technology budget exceeding $18 billion, including significant allocations to AI and machine learning, JPMorgan Chase ranks at the top of the Evident AI Index for AI maturity in banking.
The bank’s AI strategy encompasses over 450 use cases in production, spanning back-office automation, client services, and risk mitigation, with plans to expand to 1,000 by 2026. This includes proprietary platforms and collaborations with AI leaders like OpenAI and Anthropic. JPMorgan Chase’s approach emphasizes data security, employee training, and measurable ROI, positioning it as a model for AI integration in financial services.
This article explores two business use cases of AI at JPMorgan Chase:
Enhancing employee productivity and efficiency with generative AI (GenAI): deploying a proprietary LLM Suite to automate routine tasks like drafting documents and generating insights, boosting workforce efficiency across divisions.
Leveraging Machine Learning for Real-Time Fraud Detection: analyzing transaction patterns and preventing fraudulent activities via the OmniAI platform in order to reduce losses and improve security.
Enhancing Employee Productivity and Efficiency with GenAI
Financial institutions like JPMorgan Chase face mounting pressure to optimize operations amid rising costs and talent shortages. According to a McKinsey report, generative AI could add $200-340 billion in annual value to the banking sector by automating knowledge work, such as report generation and data analysis, potentially increasing productivity by 30-50% in targeted areas. However, challenges hamper this potential. They include ensuring data privacy, integrating AI with legacy systems, and training employees to use these tools effectively without introducing errors or biases.
At JPMorgan Chase, the pressure to optimize operations is amplified by the company’s scale: with over 300,000 employees handling complex workflows across compliance, marketing, and advisory services, manual processes were time-consuming and prone to inefficiencies. The bank reorganized key business units to accelerate its data and AI strategy, emphasizing modernizing systems and data infrastructure to enhance efficiency and innovation.
In a 2023 earnings call, CFO Jeremy Barnum characterized the firm’s AI deployment as measured and focused on strengthening core data and technology foundations, reinforcing investments in scalable platforms to support long-term competitiveness. described the firm’s AI rollout as disciplined in building foundational capabilities. Prior to AI adoption, employees spent hours on repetitive tasks such as drafting performance reviews or summarizing research, leaving little time for high-value activities.
To address these issues, JPMorgan Chase developed LLM Suite, a proprietary generative AI platform launched in the summer of 2024. This model-agnostic tool integrates large language models from providers such as OpenAI and Anthropic, connecting to the bank’s internal databases and applications to deliver secure, customized outputs.
LLM Suite supports tasks such as:
idea generation
content drafting
workflow automation
According to this Forbes article, the system updates every eight weeks to incorporate new capabilities, including adding more connections to the bank’s internal databases. The same article describes some specific examples of productivity gains. For example, LLM Suite enables an investment banker in the firm to generate a presentation deck in about 30 seconds that previously took a junior analyst hours to complete. Also, the article explains that Chief Analytics Officer Derek Waldron demonstrated the platform’s capabilities by asking it to prepare a 5-page presentation for a meeting with the CEO and CFO of a major tech company, and it did so nearly instantly.
The AI rollout began with an opt-in model, fostering “healthy competition” among employees, as described by Chief Analytics Officer Derek Waldron in a McKinsey report. Specialized variants, such as Connect Coach for Private Bank advisors, provide real-time, personalized insights via natural language processing.
Implementation involved rigorous governance: initial bans on external tools such as ChatGPT ensured data security, followed by internal development focused on explainability and bias mitigation through the bank’s AI Research program. Training emphasized “learn-by-doing,” with over 200,000 employees onboarded within 8 months—about two-thirds of the workforce. The platform’s integration with tools like Microsoft 365 Copilot further enhances usability.
In the video above, Teresa Heitensenrether, Chief Data and Analytics Officer at JPMorgan Chase, explains how the company is using LLM Suite.
Outcomes have been considerable:
Employees report 30-40% efficiency gains, with AI benefits growing at a similar rate each year.
In asset and wealth management, AI has reimagined workflows, enabling advisors to serve more clients more effectively. The bank estimates up to $1.5 billion in annual value from its AI initiatives.
10-20% efficiency gains for engineering teams using AI coding assistants integrated with the platform.
Leveraging Machine Learning for Real-Time Fraud Detection
The banking industry continues to grapple with escalating fraud. A study by Juniper Research forecasts that fraud could cost financial institutions as much as $58.3 billion by 2030.
When it comes to flagging potentially fraudulent transactions, more is not always better. Traditional rule-based systems often produce high false-positive rates, up to 95% in some cases, according to a 2024 qualitative research article. The result leads to operational inefficiencies and customer friction.
A J.P. Morgan article highlights the significant issues caused by false positives. Losses from false positives account for 19% of the total cost of fraud, compared with actual fraud losses, which represent an estimated 7% of that cost.
Machine learning offers a solution by analyzing vast datasets in real time, but challenges include adapting to evolving threats, integrating with existing infrastructure, and maintaining regulatory compliance.
JPMorgan Chase, processing billions of daily transactions, faced similar hurdles: manual reviews were slow, and legacy systems struggled to handle sophisticated scams such as AI-generated deepfakes and synthetic identities. There is a pressing need for advanced fraud detection. The bank’s exposure across retail, commercial, and investment banking led to heightened risks, and the firm is well aware of the lengths fraudulent actors will go to to perpetrate their scams.
To address these risks, JPMorgan Chase built OmniAI, an enterprise-wide machine learning platform launched to standardize processes and accelerate AI adoption across all lines of business, including fraud detection.
Screenshot from AWS re:Invent 2020: A day in the life of a machine learning data scientist at JPMorgan Chase
The above high-level system architecture diagram illustrates the bridge between OmniAI, its users, and governance and compliance controls to AWS cloud services, showing how it achieves secure, managed access to computational tools for data scientists, engineers, and other employees at JPMorgan.
OmniAI uses advanced algorithms to monitor transactions in real time, analyzing patterns, behavioral data, and anomalies across millions of data points. While OmniAI supports a wide range of use cases, one of its most impactful applications has been in fraud detection.
A 2024 article in the International Journal of Scientific Research and Engineering Trends indicates that the bank’s AI-based fraud prediction system saves $250 million annually.
Fraud detection enabled by OmniAI has produced measurable security and operational improvements at JPMorgan, including:
Increased savings from strategic risk management in the form of cost avoidance
Significant loss prevention in excess of $1 billion
Enhanced accuracy and reduced false positives
Operational efficiency gains resulting from accelerated insights