1Password lets AI agents log in without seeing passwords

Password manager 1Password has launched 1Password for Agents, a tool that lets users authorize AI agents to sign in to personal accounts without giving the agents access to their passwords or other credentials.

The system connects the agent, the user and 1Password. When an agent tries to log in on a user’s behalf, it sends a unique digital key to 1Password to confirm its identity. The user receives an alert and can approve the request. 1Password then enters the credentials for the agent, allowing it to sign in and complete the requested task.

Users can choose among agents including Claude Duet from Anthropic, OpenAI’s Dots or ChatGPT Work, Grok Bot, Meta’s Muse and agents from AI assistant startup Town. 1Password says it is working with other companies to expand the service. The company said in a blog post that the tool is intended to let people delegate tasks while keeping control of the credentials those tasks require.

The launch comes as Axios reports that U.S. banks have begun detecting agent activity in customer accounts. The source says this does not necessarily mean agents are acting without permission; consumers are increasingly allowing automated services to act on their behalf, including on financial matters. However, bankers told Axios they find it difficult to determine whether such activity is tied to user requests. A Chase spokesperson told Axios that customers “should not give their usernames and passwords to anyone.”