The useful question is what changes for users, developers or buyers, and whether the announcement stays industry context or becomes something people can actually use.

Agentic AI systems are evolving from tools that assist humans into independent actors capable of executing complex tasks without direct oversight. In retail, AI agents will soon research, negotiate and complete purchases on behalf of consumers, fundamentally altering traditional shopper-merchant dynamics. Brands will need to adapt by optimizing product data for machine readability rather than human search engines, as visibility increasingly depends on generative engine optimization (GEO) and standardized protocols like the universal commerce protocol (UCP).

Within enterprise operations, agentic AI is transforming supply chains and manufacturing by replacing static analytics with autonomous decision-making. Prescriptive engines driven by AI agents will adjust production schedules, reroute shipments based on real-time conditions and negotiate contracts without human input. Warehouse execution systems (WES) and technologies like RFID and spatial computing will enable autonomous inventory management, workforce optimization and loss prevention in stores, supported by real-time digital twins of physical operations.

Regulatory changes, particularly the EU’s digital product passport (DPP) mandate, will require verifiable digital records of products’ lifecycle, sustainability and composition. Compliance will demand unified, governed data architectures that integrate consumer profiles, inventory, pricing and supply chain information into a single source of truth. Cloud-native, event-driven platforms are essential to provide agents with near real-time data access, enabling sub-second decision-making in negotiations and operations.

To operationalize agentic AI, enterprises must build a robust platform capable of deploying, governing and scaling autonomous agents across business functions. This requires a unified semantic layer to standardize business logic, GEO-ready product catalogs enriched with detailed attributes, and knowledge graphs that map relationships between data sets. The platform must also enforce human oversight through human-on-the-loop (HOL) governance while enabling agents to act autonomously within defined boundaries.