The news: JPMorgan Chase spends $2 billion per year on AI and finds an equal amount of cost savings as a result, said CEO Jamie Dimon in a recent Bloomberg TV interview.

The strategy: During its April investor day, JPMorgan forecast spending $18 billion on technology in 2025, noting initiatives that include:

• Constant transformation: In Dimon’s latest letter to shareholders, he wrote that “constant transformation in technology and [system migrations] means you can’t have stop-start strategies almost anywhere.”
• Focus on AI talent: The bank has 2,000 people “working on AI.” That helped push it to the top of intelligence platform Evident’s recent financial services AI index, narrowly ahead of Capital One.
• Modern infrastructure: About 65% of JPMorgan’s applications now run primarily in the cloud, and 80% operate on "modern infrastructure"—up 15 and 5 percentage points, respectively, from 2023. Core transformation moves have made headlines.

JPMorgan's LLM Suite illustrates these principles at work. Debuted in 2024, it reportedly has 200,000 users out of the banks' roughly 317,000 employees. It has also noted:

• Over 40,000 of its software engineers use AI coding assistants.
• Customer service representatives increasingly use AI in call centers to resolve issues more efficiently and effectively.

How we got here: JPMorgan is one of few financial institutions with the institutional resources to support the independent advancement of AI. It has not only the necessary finances ($177.56 billion in annual total net revenues to draw from), but also the platform modernization and scale.

Our take: JPMorgan is increasing the gap between the haves and have nots in bank technology. AI development in financial services, supported by modern platforms, is outrunning nearly everyone.