Skip to content

AI VERSE NEWS

Menu
  • Home
  • AI Tools
  • AI Breaking News
  • AI Tools & Products
  • Google AI
  • Privacy Policy
  • Terms of Use
  • contact
Menu

OpenAI’s new ChatGPT Wallet lets AI shop for you. What are the risks?

Posted on by admin

OpenAI recently introduced ChatGPT Wallet, a new payment tool that lets AI agents make purchases on a user’s behalf. This company built it in partnership with American Express, Mastercard, and Visa. 

Eligible users add their payment information once, and then their AI agent can take it from there and suggest purchases that users can approve with an added security check. 

This launch comes at a time when more consumers are adopting AI across various areas of their personal and professional lives to manage their schedules, comparison shop, seek recommendations, advice, and more. 

However, experts warn that convenience can come with added risks. “The first thing to understand is that this is a frontier technology and is probably not appropriate for most people today,” said Identity Theft Resource Center president James E. Lee. “It’s better suited to early adopters who have a high level of risk tolerance and understanding of AI technology.”

“With any personal agent, you need to be certain you understand how your personal information is going to be used, where it will be stored, what kind of permissions are required to approve a transaction, how you can revoke permissions to access your information, and the process for disputing any transaction made on your behalf,” Lee said.

Keep track of your finances for free with My Money

How ChatGPT Wallet works and who is eligible to use it

In late September, OpenAI began rolling out ChatGPT Dots to eligible ChatGPT users. Dots are what the company calls “always-on agents” that can work in the background even when you’re not actively using ChatGPT to manage tasks, make progress on projects or research, and work across apps. 

Dots can now complete purchases. 

Before each purchase, users review the items, total, and selected card, then approve it with Face ID or a fingerprint scan. Each approval creates a clear record of the purchase. 

After the user approves the purchase, the payment network provides a payment token containing a card number, security code, and expiration date in place of your personal card information, allowing the payment to be completed without ChatGPT Wallet sharing actual card details with the business or agent.

According to OpenAI’s official release, the company expects this product to evolve over time, allowing users to offload purchases entirely to their AI agent, with parameters such as spending limits, purchase categories, and more. 

ChatGPT Wallet is initially available to ChatGPT Pro subscribers in the U.S. who are 18 and older and have eligible U.S.-issued cards. 

Open AI CEO Sam Altman speaks at OpenAI DevDay 2026 conference, Tuesday, Sept. 29, 2026, in San Francisco. (AP Photo/Jeff Chiu)
Open AI CEO Sam Altman speaks at OpenAI DevDay 2026 conference, Tuesday, Sept. 29, 2026, in San Francisco. (AP Photo/Jeff Chiu) · AP Photo/Jeff Chiu

AI can simplify money management, but it can carry security risks 

Reviewing a monthly budget, building a debt pay-off plan, or comparing purchase options can be tedious, and AI tools can take some of the guesswork out of those tasks. Many Americans are turning to them.

According to TD’s second annual U.S. AI Insights Report, 78% of Americans say they use AI-enabled tools in their daily lives, and more than half say they use AI to help them make financial management decisions. 

However, widespread adoption of AI could also potentially open the door for oversharing of sensitive information. 

A 2024 Consumer Privacy Survey by Cisco asked users what information they share with generative AI tools. It found that 37% of respondents have submitted health information, 36% have entered work information, and 24% to 29% of users have provided financial information, account numbers, and religion/ethnicity. 

The information isn’t guaranteed to stay private. Some developers may allow humans to review users’ conversations with their AI agents for training purposes. Users should read the terms of their accounts carefully and understand how their information is used across various AI platforms, and opt out of any settings they’re not comfortable with.

As AI becomes more deeply embedded in everyday life, there are steps you can take to protect your sensitive information: 

  • Opting out of training models: You can prevent certain platforms from using your conversations to inform future conversations by opting out of this in your data control settings. 

  • Never upload files that contain sensitive information: Avoid uploading personal documents that include sensitive information such as your Social Security number, credit card numbers, bank account numbers, medical records, or any other confidential information. 

  • Using temporary or incognito chats: Depending on the AI platform, you may be able to use a temporary or incognito chat that has no memory of past chats. These incognito chats are not stored in your chat history and are not used to train models.

  • Reviewing your account settings regularly: Read the terms and conditions of your account to better understand how your information is being collected and stored. You may even want to review these terms periodically so you don’t miss any privacy policy changes and can adjust your privacy settings as needed. 

Experts also urge consumers to weigh their options carefully before allowing AI to make spending decisions for you. 

“In many ways, giving your personal and financial information to an AI agent to act on your behalf is like giving a total stranger your information,” Lee said. “You probably wouldn’t do that without doing a lot of due diligence and, even then, you would put a lot of restrictions on how that information could be used. “

Lee says users should carefully consider whether the convenience and novelty of having an AI agent make purchases is worth the risk.

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

  • Nvidia reports bumper results as firm says it’s not reinstating Jensen Huang
  • Tech outages live updates: Microsoft, banks and FAA
  • Trump administration to name Nvidia, other AI firms as US ‘national champions’, WSJ reports
  • A Man and His Muse AI Agent Sue Meta
  • Microsoft’s AI pitch at Ignite: Cheaper models as business ‘token spend is going through the roof’
©2026 AI VERSE NEWS | Design: Newspaperly WordPress Theme