Artificial intelligence is reshaping commerce, with consumers already using AI to search, compare and guide purchases. The next phase is Agentic commerce which enables AI agents to act on a person’s behalf to complete transactions. Critically, trust, consent and security must be built in from the start.
In January, Mastercard announced Australia’s first fully authorised agentic payment, demonstrating how AI-driven commerce can work safely in the real world. Through Mastercard Agent Pay, an AI agent is able to initiate a payment only after verifying the agent, capturing clear user intent, and receiving explicit customer consent, with full transparency at every step.
In the demonstrations, while planning a friend’s trip to Thredbo or buying movie tickets for the family, instead of navigating multiple sites, comparing options and managing payments manually, an AI agent curated the best option based on the customer’s preferences and budget, then completed the transaction securely on their behalf, all within defined permissions and with real-time oversight.
This technology is here today, and trusted payments infrastructure is at the heart of it. Mastercard’s tokenised, secure network ensures agentic commerce is governed, auditable and safe for consumers, businesses and banks alike. They value and invest in AI capability development, from foundational literacy through to advanced technical and role-specific training. As Australia looks to move AI from experimentation to practical deployment, trust and security will determine how quickly agentic commerce can scale responsibly.