The customer engagement center market has moved past the clean demo of AI answering a question. The more useful test now is whether CX platforms can connect intent, data, workflow, inventory, service policy, and human approval without making the operating model harder to manage.
That is the thread running through Adobe, Orange France, Zendesk, and Oracle this week. Each story points to a different part of the same buyer question: can the technology actually complete work inside the enterprise, or does it simply make the conversation around that work sound more intelligent?
Adobe Pushes CX Coworker Into Enterprise AI Platforms
Adobe has made its CX Enterprise Coworker available inside major AI platforms, including ChatGPT as a plugin, Claude, Databricks Genie, Gemini for business, and Microsoft Copilot.
Availability in Amazon Quick, IBM, and Snowflake is expected to follow, according to Adobe.
The strategic point is Adobe is trying to put its customer experience orchestration layer into the AI environments where enterprise teams already work, rather than forcing marketers to return to a separate interface every time they need to plan, analyze, or act.
Adobe says CX Enterprise Coworker supports autonomous customer experience orchestration workflows across Adobe CX Enterprise, AI platforms, and enterprise systems with human oversight. It is built on open standards including Model Context Protocol and Agent2Agent.
The product claims matter because CX work usually crosses data, content, journeys, approvals, and channel execution. A marketing assistant that drafts copy is useful. A system that can analyze performance, forecast audiences, resolve journey conflicts, and coordinate work across partner systems moves into a different category of operational dependence.
Adobe is also adding pre-built Model Context Protocol server integrations with partners including Amazon Ads, Databricks Genie, Demandbase, Meta campaign infrastructure, Microsoft Work IQ, RainFocus, and Snowflake Cortex AI.
That connector strategy is where the announcement takes on a more serious tone. Enterprise buyers need controlled access to the systems where customer signals, content, budgets, and execution decisions already live. Anjul Bhambhri, Senior Vice President of Engineering, Customer Experience Orchestration at Adobe, framed the problem as one of context rather than raw AI capability:
“Today, marketers often use AI tools in isolation, without the organizational context and domain intelligence needed to personalize every stage of the customer lifecycle.”
The open question is how much autonomy enterprises will really allow across those connected systems. Adobe’s partner ecosystem, including systems integrators and agencies, gives the proposition scale. It also raises the familiar deployment challenge: every useful integration adds another place where permissions, governance, data quality, and accountability have to hold up.
Orange France Rebuilds Fulfillment Around Real-Time Execution
The Orange France story is less loudly “AI,” but arguably just as important for CX leaders.
Kbrw and Accenture have been selected by Orange France to modernize order management and inventory execution across online channels, retail stores, and B2B operations.
The first phase enables customers to receive more accurate product availability and more reliable delivery dates when buying smartphones, SIM cards, and telecom equipment. It also supports pre-ordering and inventory reservation.
Orange France will extend the system to replenishment orders for retail stores next. B2B and enterprise orders are expected to follow in 2027, with after-sales services also planned over time.
The architecture is the useful detail. Orange France is separating transactional functions from operational functions. Its Enterprise Resource Planning system remains the main record for financial and transactional processes, while Kbrw’s Order Management System manages orders, inventory, and fulfillment in real time.
That distinction should resonate with CX teams. Customer experience often breaks at the handoff between a promise made in the channel and the operational system that must fulfill it.
A website can show a delivery estimate. A service agent can reassure a customer. Neither interaction creates trust if the inventory, warehouse, logistics, and returns processes behind the promise are fragmented. Lionel Benezech, Supply Chain Director at Orange France, described the execution goal clearly:
“This new program gives us a single and consistent view of products that are in stock, in transit, or expected to arrive with projected inventory, helping us improve inventory planning and allocation across our sales channels.”
Personalization and service recovery depend on operational truth. If teams cannot see stock availability, delivery constraints, reservations, returns, and repair flows accurately, AI will only make unreliable promises faster.
Zendesk Uses Gartner Recognition to Reframe Service AI
Zendesk has been named a Leader in the 2026 Gartner Magic Quadrant for CRM Customer Engagement Center for the second consecutive year.