Stripe Gives AI Agents Dedicated Link Balances

Nov 1, 2026
Stripe launches Link balance, which lets consumers fund a balance and set a per purchase limit so AI agents can buy without approval each time.

Stripe has launched Link balance, which lets consumers fund a dedicated balance and set a per purchase limit for their AI agents, Stripe Crypto said in a post on 27 October. The product starts with Instinct, an AI personal assistant. Purchases within the limit go through without approval each time; larger transactions still need the consumer's approval.

Consumers add money with a debit card. Stripe Crypto Onramp converts the dollars into a dollar denominated stablecoin on Tempo, and the funds sit in a non custodial wallet powered by Privy, which handles key management and transaction signing. Consumers see a balance in dollars. Stripe says this is the first consumer launch of Stripe Financial Accounts backed by Privy's non custodial wallets.

When an agent buys something, Stripe Issuing provides a virtual debit card backed by the Link balance. Merchants accept a normal card payment through their existing checkout and do not need to support stablecoins.

Until now, consumers had to approve every purchase an agent made through Link Agent Wallet, which agents such as Muse, Grok Bot and Instinct already use. According to Stripe, agentic purchases made with Link grew 38 times over the past month.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Finance AI Weekly, Retail & Ecommerce AI Weekly or Daily AI Brief.

Also, consider following us on social media:

Free newsletter

Retail & Ecommerce AI Weekly

The weekly on AI in retail and ecommerce: shopping agents, online stores and marketplaces, checkout, and AI inside physical stores.

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more
Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.