Multiple American banks require customers to bear the risks of AI agency operations and theft themselves
According to TechCrunch, Citibank updated its online user agreement this week, stating that customers must assume responsibility for activities conducted by third-party AI agents after authorizing them to access their accounts. If AI tools leak account credentials or are hacked, customers must also bear the responsibility for operations carried out by hackers through their accounts. A Citibank spokesperson stated that the update aims to clarify the risks associated with customers sharing banking credentials or authorizing third-party tools to access their accounts.
Last month, JPMorgan Chase and Bank of America announced similar policies, and PNC Bank adjusted its terms on November 18. Citizens Bank also sent emails to customers this week, reminding them that they are responsible for operations performed by AI assistants. The Citibank agreement also states that customers using AI agents provided by Citibank do not assume additional liability.






