The next stage of automated payments may depend less on how fast a transfer settles and more on proving who authorized it. On October 20, Visa, BILL, and Billtrust announced a live pilot allowing trusted AI agents to initiate and complete approved buyer-invoice payments through supplier portals.
The project utilizes Visa’s Intelligent Commerce tools, which include the Trusted Agent Protocol, to authenticate software agents to portals that historically expected a human user. The participating companies emphasize the preservation of supplier oversight and controls. This initiative functions as a limited business-payments pilot rather than an announcement that autonomous systems hold open-ended spending authority on behalf of users.
A close parallel exists for cryptocurrency wallets. Although an agent can technically sign and broadcast transfers currently, complex questions remain regarding budgets, permissions, revocation, audit trails, and accountability when an agent makes an error. Businesses must resolve these exact issues before automatic onchain commerce can become routine.
While Visa’s experiment is not a public-blockchain settlement launch, its relevance to the crypto sector lies in the permission layer. Regardless of the rail eventually carrying an automated payment, businesses require evidence that the initiating software was known, constrained, and operating within specific instructions. This underlying security challenge is evolving into a distinct commercial product category.
