Agentic commerce and enterprise AI: What businesses must know now

The rise of agentic commerce and enterprise AI is no longer hypothetical. Mastercard says its network already saw an agent complete a purchase. Therefore, payments, marketplaces, CRM vendors and frontline sales teams must adapt fast. This post pulls together five recent signals from Mastercard, Gartner, eBay, AWS and Sysco. Additionally, it explains what each development means for business leaders. The aim is simple: give clear context, practical impact and a short view on what comes next.

## Mastercard: Why a payments network declaring "agentic commerce is here" matters

Mastercard used its October earnings call to mark a turning point. The company said its network recorded the first agentic transaction this quarter. Additionally, Mastercard announced that Agent Pay — a platform to let AI agents complete purchases — is rolling out in the U.S. now. This is important because payments networks are the plumbing behind every online purchase. Therefore, when a major card network signals it can process agent-driven buys, it lowers a big barrier to adoption.

For merchants and platforms, the implication is clear. Agentic shopping — where an AI agent researches, selects and pays for items on behalf of a person — needs reliable, secure payment flows. Mastercard moving from proof to live transactions shows those flows are becoming real. However, this also raises questions. Merchant integration, fraud checks and user consent flows will need redesign. Additionally, payment firms and banks will have to build rules for agent authorization and liability.

In short, Mastercard’s move accelerates industry readiness. Therefore, companies that sell online must start planning for AI agents interacting with their checkout. Otherwise, they risk missing a fast-shifting buyer behavior and new revenue paths.

Source: Digital Commerce 360

Gartner’s CRM CEC verdict: Intelligent orchestration and agentic commerce and enterprise AI

Gartner’s 2025 Magic Quadrant for CRM Customer Engagement Center (CEC) paints a bigger picture. The market is shifting from simple digital engagement to intelligent orchestration. Therefore, customer service platforms are being rethought around AI that coordinates channels, data and actions. Gartner’s analysis signals that vendor roadmaps must account for agentic workflows and automated decision-making.

For enterprises, the takeaway is practical. CRM and contact center teams can no longer treat AI as a nice-to-have feature. Instead, they must view it as the layer that will orchestrate customer journeys across marketing, service and sales. However, orchestration is hard. It demands cleaner data, common APIs and governance to ensure actions are accurate and compliant. Additionally, selecting a platform now involves testing its capacity to route tasks to human agents, automated systems and third-party AI agents.

This shift also changes buying criteria. Enterprises should prioritize vendors that show a strategy for safe automation, transparent decision logs and easy integration with payments and marketplaces. As agentic commerce grows, CRM systems that can handle automated purchasing requests, confirmations and issue resolution will add real customer value.

Source: CX Today

eBay’s tests: Marketplaces putting agentic commerce and enterprise AI into practice

eBay is experimenting with agentic shopping and working with OpenAI. Over recent years, eBay has increased its use of generative AI. For example, it has used AI to improve listings with a tool the company calls “magical listing.” Now, the marketplace tests how agentic agents can interact with buyers and sellers to complete transactions or improve listings automatically.

This matters because marketplaces are uniquely poised to benefit. They control catalog data, listings, pricing and dispute processes. Therefore, a marketplace that supports agentic flows can streamline buying for repeat customers, or let agents negotiate and complete purchases for busy buyers. Additionally, seller experiences can improve if AI helps create better listings and respond to buyer queries faster.

Yet, marketplaces face trade-offs. Automated agent actions need clear seller consent and fair dispute handling. Furthermore, trust signals and verification must be robust to avoid scams. eBay’s testing with a major AI provider shows how large marketplaces are taking a careful step toward agentic commerce. As a result, both buyers and sellers should watch how marketplaces introduce these features and what controls they offer.

Source: Digital Commerce 360

AWS warning: Why caution around third‑party agents matters for agentic commerce and enterprise AI

Amazon Web Services recently suggested that many third‑party agents are not yet delivering acceptable customer experiences. In its quarterly briefing, AWS said most current third-party agents lack the capabilities required for a reliable agentic commerce experience. Therefore, enterprises should be cautious when outsourcing agent functions to third parties.

This warning is critical for procurement and operations teams. First, it tells buyers to set strict evaluation criteria for any agent vendor. These criteria should include accuracy, escalation paths, audit logs and measurable customer satisfaction. Second, it means businesses need contingency plans. For example, automated agents should have clear handoffs to humans when confidence is low.

However, caution does not mean stopping. Instead, it argues for staged deployments. Enterprises can pilot agents in low-risk areas, gather performance data and scale when outcomes are proven. Additionally, vendors and integrators must improve agent training, context handling and safety guardrails. As agentic commerce grows, reliable user experiences will separate long-term winners from short-lived experiments.

Source: CX Today

Sysco’s playbook: Frontline adoption shows agentic commerce and enterprise AI impact on sales

Sysco put AI at the center of its sales rebound. The company reported that roughly 90% of its sales consultants use AI tools. Therefore, this is a strong example of enterprise teams adopting AI not as a novelty but as a day-to-day assistant. For companies that sell complex goods or need frequent customer contact, AI can boost productivity and help close deals.

The Sysco case shows two practical lessons. First, frontline adoption scales faster when tools solve real, immediate problems — like drafting proposals or finding product matches. Second, training and change management matter. Sysco’s large uptake suggests the company invested in making tools easy to use and aligned them with sales incentives.

For leaders, the implication is straightforward. If agentic commerce models reach customers, sales teams will need new roles and skills. Additionally, businesses must decide where agents can act autonomously and where human approval is required. Finally, measured rollout and performance tracking help ensure AI improves revenue without harming customer relationships.

Source: Digital Commerce 360

Final Reflection: Connecting the signals into a practical roadmap

Taken together, these reports trace a clear arc. Mastercard’s live agentic transaction shows the payment rails can handle agent-driven buys. Gartner’s Magic Quadrant emphasizes that CRM and contact centers need intelligent orchestration to manage automated interactions. eBay’s experiments show marketplaces are testbeds for agentic flows. However, AWS’s caution reminds us that many third-party agents still fall short on customer experience. Finally, Sysco demonstrates how broad internal adoption can drive results when tools are practical and easy to use.

Therefore, leaders should treat agentic commerce as an urgent but manageable transformation. Start with pilots that connect trusted data, payments and escalation paths. Additionally, insist on vendor transparency, performance metrics and clear customer consent. With careful rollout, businesses can capture the convenience and efficiency of agent-driven commerce while protecting customers and brand trust. The opportunity is real. The next challenge is to build the governance and integrations that make it safe and scalable.