JPMorgan is giving its AI coding tool to more of its employees

JPMorganChase is giving Claude Code, an AI tool that helps write and check software, to more of its employees. It is no longer just for engineers. Product and risk teams are using it too, according to a post on the Claude engineering blog. The bank says work in its Commercial and Investment Bank is reaching production faster.

What changed

The rollout now reaches people closer to the business side. Product teams use Claude Code to turn requirements directly into engineering backlogs, which are to-do lists for software teams. Other teams use it for risk and controls assessments and for program planning. Infrastructure and operations teams use it for monitoring, alerting and root cause analysis, which means finding out why something broke.

Engineers are also building their own agents, meaning AI helpers that carry out tasks. Examples include agents that review code changes and so-called self-healing agents. The firm's security team uses the AI models to find and fix vulnerabilities faster. Teams are also modernizing older legacy systems.

Wealth management and clients

In Asset & Wealth Management, people with deep business insight now work alongside technology and product experts in small teams. Claude Code is part of their engineering workflow. The firm says these teams are moving faster while putting every change through more rigorous testing before it reaches clients.

Last month, J.P. Morgan Asset Management announced a connector that brings 55ip's ActiveTax Technology into Claude for Financial Advisors. It puts tax-aware portfolio transition analysis into advisors' daily work.

The safety setup

Claude Code runs in controlled environments, including Devspace, the firm's containerized development platform on its own infrastructure. The intent is to stop agents from getting direct access to employee credentials or internal systems. Access widens in stages. New Claude models go through the firm's own approval process before anyone there can use them.

• Thousands of automated tests and approval gates check every change.
• An automated gate returns a hard pass or halt, and the pipeline enforces it.
• Claude cannot sign off on its own work.
• People stay accountable for every production change.

What is not known

The post gives no numbers. It does not say how many employees now have access, how much faster work moves, or how much money the firm expects to save. Those claims come from the company and the AI maker, and no outside check is included.

What it means for your business

The pattern is the useful part. The bank widened access in steps, kept the AI away from passwords and core systems, and put automated checks between the AI and anything that goes live. A person still owns each change. Smaller companies can borrow the same idea: let AI speed up the work, but keep sign-off with people and tests.