top of page
OutSystems-business-transformation-with-gen-ai-ad-300x600.jpg
OutSystems-business-transformation-with-gen-ai-ad-728x90.jpg
TechNewsHub_Strip_v1.jpg

LATEST NEWS

OpenAI launches ChatGPT Wallet to enable secure purchases through AI agents

Marijan Hassan - Tech Journalist
11 minutes ago
2 min read

OpenAI has introduced ChatGPT Wallet, a dedicated payment infrastructure allowing its autonomous AI agents, known as dots, to complete online purchases and bookings on a user's behalf. Developed in partnership with American Express, Mastercard, and Visa, the system relies on network-issued payment tokens and mandatory biometric checks to let agents execute checkouts without ever exposing real credit card numbers to the AI model or merchants.


Editorial credit: Primakov / Shutterstock
Editorial credit: Primakov / Shutterstock

The feature began rolling out gradually on November 13 to eligible ChatGPT Pro subscribers aged 18 and older in the United States with a U.S. billing address.


Tokenized Transactions and Mandatory Biometric Guardrails

To address widespread consumer and financial sector skepticism regarding autonomous AI spending, OpenAI designed ChatGPT Wallet around human-in-the-loop verification and card-network tokenization:

  • Zero Card Exposure: Users enroll an eligible U.S.-issued credit card (debit cards are currently unsupported) once via a secure enrollment form. When a transaction is triggered, the card network generates a unique substitute payment token, complete with a virtual card number, security code, and expiration date, ensuring neither the merchant nor the AI agent sees the physical card details.

  • Biometric Approval Checks: During the initial rollout, autonomous spending without human confirmation is disabled. Before every purchase, users must review the exact items, merchant, total cost, and selected card, then authorize the payment via device biometrics such as Face ID or Touch ID.

  • Preserved Card Protections: Because transactions route directly through standard payment networks, users retain existing credit card rewards programs, fraud monitoring, and purchase dispute protections.


OpenAI confirmed that future updates will allow users to pre-approve autonomous purchases within strict user-defined guardrails, such as spending caps, approved merchant categories, and specific time windows.


Navigating Merchant Friction and Bot Blockades

The launch lands amid growing friction between AI developers and major e-commerce platforms. Several leading retailers and financial institutions, including Amazon and major U.S. banks, have actively deployed bot defenses to block third-party AI shopping agents from scraping inventory or automating checkouts.


To encourage retailer adoption, OpenAI simultaneously published a Commerce Partner Playbook outlining technical integrations for online stores:

  • Signed Agent Authentication: Guidance on configuring Content Delivery Networks (CDNs) and web application firewalls to recognize signed ChatGPT agent traffic via Web Bot Auth rather than blocking it as malicious bot activity.

  • Fraud & Payment Stack Alignment: Partnerships with payment processors and fraud-prevention providers, including Stripe, Adyen, Checkout.com, Forter, Riskified, Sardine, and Signifyd, to prevent legitimate agentic transactions from triggering false fraud declines.

  • Direct Storefront Feeds & WebMCP: Tools enabling merchants to claim their business profiles inside ChatGPT, sync live pricing and inventory feeds, and deploy WebMCP integrations so agents can manage shopping carts directly.


By pairing major card-network tokenization with strict biometric sign-offs, OpenAI is attempting to bridge the trust gap required to turn ChatGPT's 1.2 billion weekly users from product researchers into direct agent-assisted buyers.

wasabi.png
Gamma_300x600.jpg
paypal.png
bottom of page