2 min read
Add as a preferred source on Google

Wells Fargo Sets Customer Responsibility Terms for AI Agent Use

The agreement is marked effective Nov. 24, 2026. Its agent section applies “if and when” Wells Fargo permits agent use.

A mechanical hand reaches toward a smartphone beside glass / TokenPost.ai
A mechanical hand reaches toward a smartphone beside glass / TokenPost.ai

Wells Fargo’s upcoming online banking agreement assigns customers responsibility for AI agent activity and gives the bank authority over access, putting formal terms in place before a Nov. 24 effective date.

When a customer provides account credentials to an agent or otherwise authorizes it to act on the customer’s behalf, the agreement treats the agent’s instructions, actions and transactions as initiated and authorized by the customer without further verification. Applicable law or regulation may provide an exception.

The agreement’s overview says customer responsibilities apply “if and when we permit the use” of agents. Section 15 also covers a customer who uses, allows, enables or authorizes an agent to access or interact with an account or service. The terms do not identify which agents or access methods Wells Fargo currently permits.

Customers must monitor agents. The agreement separately assigns customers responsibility for agents’ actions, errors, unintended actions and related results, subject to limits under law or regulation. Its definition includes software, applications, programs or other technology that acts for a person or entity, whether autonomously, semi-autonomously or non-autonomously.

Wells Fargo reserves the right to require customers to register their agents and to change or limit access methods. It may also prohibit, restrict, suspend or terminate an agent’s access.

Other major U.S. banks have described different controls. JPMorganChase CEO Jamie Dimon said Oct. 13 that the bank would not allow outside agents unrestricted access, citing risks to clients. CFO Jeremy Barnum said some services that enable money movement use credential sharing, and the bank blocks those services when it can. Bank of America CEO Brian Moynihan said Oct. 14 that the bank is working with agent providers to keep human oversight before money leaves an account. A Citi spokesperson said third-party agents cannot log into customer accounts or transact directly on its platforms, while certain Citi-approved open banking providers can provide limited read-only access.

The agreement places customer obligations and Wells Fargo’s access rights in the same section, while leaving permitted agent types unspecified. It is marked effective Nov. 24, 2026.

Simon Yoon

Reporter

Simon Yoon reports on blockchain technology for TokenPost. Send corrections or tips to info@tokenpost.com.

Loading…