When a shopper authorizes an AI agent to buy on their behalf and later disputes the charge, who eats the loss? Chargebacks911, a Tampa-based dispute-management company, has an answer: it will. The firm launched what it calls the industry's first indemnification product covering chargebacks on purchases made by AI agents, shipped October 26, 2026 following its Money20/20 USA presence.
Highlights#
The product is a bolt-on, not a rip-and-replace. Merchants keep their existing fraud filters and checkout stack; Chargebacks911 sits alongside them, identifies transactions involving AI agents, and indemnifies qualifying chargebacks tied to those purchases.
Detection leans on the company's own turf: AI analysis, behavioral intelligence, and years of global dispute data are combined to spot purchasing patterns characteristic of agents and to decide which transactions are eligible for coverage.
CEO Monica Eaton framed the bet in blunt terms. Everyone tells merchants to fear AI agents, she said, but her company would rather hand them a reason not to, arguing that agentic commerce should be an opportunity rather than a transaction category merchants must accept at their own risk.
The launch lands amid an unresolved industry fight over consumer authorization, merchant liability, and how to authenticate transactions initiated by machines. Card schemes are building their own frameworks, including Visa's Trusted Agent Protocol and Mastercard's Agent Pay, while merchants worry about being stuck holding a new class of post-transaction risk.
Why it matters#
Disputes are the piece of agent commerce nobody wants to own. If machine-initiated purchases are going to scale, someone has to price and absorb the risk when a human repudiates the transaction. A third party volunteering to underwrite that exposure is a meaningful data point for escrow and reputation rails: when liability has a market, agent-to-agent trade gets a floor to stand on.